U.S. patent number 9,095,809 [Application Number 14/074,860] was granted by the patent office on 2015-08-04 for selectivation of adsorbents for gas separation.
This patent grant is currently assigned to ExxonMobil Research and Engineering Company. The grantee listed for this patent is Harry W. Deckman, Preeti Kamakoti, Peter I. Ravikovitch, Chris Yoon. Invention is credited to Harry W. Deckman, Preeti Kamakoti, Peter I. Ravikovitch, Chris Yoon.
United States Patent |
9,095,809 |
Deckman , et al. |
August 4, 2015 |
Selectivation of adsorbents for gas separation
Abstract
Systems and methods are provided for improving separation of gas
phase streams using an adsorbent, such as 8-member ring zeolite
adsorbents or DDR type zeolite adsorbents. Suitable gas phase
streams can include at least one hydrocarbon, such as methane or a
hydrocarbon containing at least one saturated carbon-carbon bond,
and at least one additional component, such as CO.sub.2 or N.sub.2.
The selectivity of the adsorbent is improved by selectivating the
adsorbent with one or more barrier compounds. The presence of the
barrier compounds is believed to alter the relative ability of
potential adsorbates to enter into and/or move within the pores of
the adsorbent.
Inventors: |
Deckman; Harry W. (Clinton,
NJ), Ravikovitch; Peter I. (Princeton, NJ), Kamakoti;
Preeti (Summit, NJ), Yoon; Chris (Asbury, NJ) |
Applicant: |
Name |
City |
State |
Country |
Type |
Deckman; Harry W.
Ravikovitch; Peter I.
Kamakoti; Preeti
Yoon; Chris |
Clinton
Princeton
Summit
Asbury |
NJ
NJ
NJ
NJ |
US
US
US
US |
|
|
Assignee: |
ExxonMobil Research and Engineering
Company (Annandale, NJ)
|
Family
ID: |
49622910 |
Appl.
No.: |
14/074,860 |
Filed: |
November 8, 2013 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20140157984 A1 |
Jun 12, 2014 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
61734012 |
Dec 6, 2012 |
|
|
|
|
61734007 |
Dec 6, 2012 |
|
|
|
|
61734010 |
Dec 6, 2012 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
B01D
53/047 (20130101); C01B 37/02 (20130101); B01D
53/02 (20130101); B01D 53/0473 (20130101); B01D
53/0462 (20130101); B01D 53/04 (20130101); C01B
39/48 (20130101); C01B 39/04 (20130101); B01J
20/3078 (20130101); B01D 69/147 (20130101); B01D
53/0423 (20130101); B01J 35/023 (20130101); B01D
2256/245 (20130101); Y02P 20/151 (20151101); Y02C
20/40 (20200801); B01D 2257/102 (20130101); B01J
35/026 (20130101); Y02C 10/08 (20130101); B01D
2253/108 (20130101); Y02P 20/152 (20151101); B01D
2257/504 (20130101) |
Current International
Class: |
B01D
53/04 (20060101); B01D 53/047 (20060101); B01D
69/14 (20060101); B01J 20/30 (20060101); C01B
39/48 (20060101); B01D 53/02 (20060101); C01B
37/02 (20060101) |
Field of
Search: |
;95/45,47,49,50,51,96,128,130,135,136,139,141,143 ;96/4,108 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Other References
Olson et al, "Light hydrocarbon sorption properties of pure silica
Si-CHA and ITQ-3 and high silica ZSM-58", Microporous and
Mesoporous Materials, 2004, pp. 27-33, vol. 67, Science Direct,
Elsevier Inc. cited by applicant .
Zheng et al, "Synthesis, characterization, and modification of DDR
membranes grown on alpha-alumina supports", Journal of Materials
Science, 2008, pp. 2499-2502, vol. 43, Springer Science+Business
Media, LLC. cited by applicant .
Den Exter et al, "Separation of Permanent Gases on the All-Silica
8-Ring Clathrasil DD3R", Zeolites and Related Microporous
Materials: State of the Art 1994, Studies in Surface Science and
Catalysis, 1994, pp. 1159-1166, vol. 84, Elsevier Science B.V.
cited by applicant .
Tomita et al, "Gas separation characteristics of DDR type zeolite
membrane," Microporous and Mesoporous Materials, 2004, pp. 71-75,
vol. 68, Science Direct, Elsievier Inc. cited by applicant .
Potapova, "Synthesis and characterization of the DDR type zeolite"
(Master Thesis), Lulea University of Technology, Department of
Chemical Engineering and Geosciences, Division of Chemical
Technology, Jun. 2007. cited by applicant .
Himeno et al., "Methane and Carbon Dioxide Adsorption on the
All-Silica DD3R Zeolite", ZMPC2006, p. 2036 (2006). cited by
applicant .
Himeno et al., "Characterization and selectivity for methane and
carbon dioxide adsorption on the all-silica DD3R zeolite",
Microporous and Mesoporous Materials, Jan. 5, 2007, pp. 62-69, vol.
98, issues 1-3, ScienceDirect, Elsevier. cited by applicant .
International Search Report with Written Opinion for
PCT/US2013/069060 dated Apr. 8, 2014. cited by applicant .
International Search Report with Written Opinion for
PCT/US2013/069080 dated Apr. 17, 2014. cited by applicant .
Ernst et al., "Hydrothermalsynthese des Zeoliths ZSM-58 and
templatfreie Synthese von Zeolith ZSM-5", Chemie Ingenieur Technik,
Jul. 1, 1991, vol. 63, No. 7, pp. 748-750. cited by applicant .
Kumita et al., "Shape selective methanol to olefins over highly
thermostable DDR catalysts", Applied Catalysis A: General, Jul. 13,
2010, vol. 391, No. 1, pp. 234-243, Elsevier. cited by applicant
.
International Search Report with Written Opinion for
PCT/US2013/069073 dated May 6, 2014. cited by applicant .
Office Action from related U.S. Appl. No. 14/074,918 dated Jun. 8,
2015. cited by applicant.
|
Primary Examiner: Greene; Jason M
Attorney, Agent or Firm: Weisberg; David M. Ward; Andrew
T.
Parent Case Text
CROSS-REFERENCE TO RELATED APPLICATIONS
This application claims the benefit of U.S. Application Ser. No.
61/734,012, filed on Dec. 6, 2012; which is incorporated by
reference herein in its entirety. This application is also related
to U.S. Application Ser. Nos. 61/734,007 and 61/734,010, also filed
on Dec. 6, 2012, and the two other co-pending U.S. utility patent
applications filed on even date herewith and claiming priority
thereto, respectively, all of which are hereby incorporated by
reference herein in their entirety.
Claims
What is claimed is:
1. A method for performing a gas separation, comprising: contacting
an adsorbent or membrane comprising an 8-member microporous
material with a barrier compound under effective conditions to
selectivate the adsorbent or membrane, the barrier compound having
a minimum molecular dimension that is at least 0.4 Angstroms
greater than the maximum dimension of the largest hard sphere that
can diffuse along any direction of the adsorbent and a maximum
dimension of about 25 Angstroms or less; wherein the effective
conditions for contacting the adsorbent or membrane with a barrier
compound comprise a temperature from about 50.degree. C. to about
350.degree. C. and a total pressure of about 100 psig (about 690
kPag) to about 2000 psig (about 13.8 MPag), wherein the barrier
compound is present either as a liquid or as a gas with a partial
pressure of the wherein compound that is at least about 10% of the
saturation vapor pressure; wherein selectivation of the adsorbent
or membrane comprises diffusion of at least one barrier compound
molecule through a pores of the 8-member microporous material;
contacting the selectivated adsorbent or membrane with an input gas
stream containing a first component and a second component to form
a first gas stream enriched in the first component relative to the
input gas stream; and collecting a second gas stream enriched in
the second component relative to the input gas stream.
2. The method of claim 1, wherein the 8-member ring microporous
material is a DDR type zeolite, Sigma-1, ZSM-58, or a combination
thereof.
3. The method of claim 1, wherein the barrier compound is a glycol,
an amine, an alcohol, an alkane, a sulfur bearing compound, or a
combination thereof, the barrier compound having a molecular weight
of at least 50 g/mol.
4. The method of claim 1, wherein the barrier compound is ethylene
glycol, triethylene glycol, methyldiethyl amine,
dimethylethylamine, dimethyl disilane, n-hexane, 2-octanol, or a
combination thereof.
5. The method of claim 1, wherein the first component is CH.sub.4
or a combination of CH.sub.4 and H.sub.2S.
6. The method of claim 1, wherein the second component is CO.sub.2,
N.sub.2, H.sub.2S, or a combination thereof.
7. The method of claim 1, wherein the effective conditions for
contacting the adsorbent or membrane with a barrier compound
comprise a temperature of at least about 150.degree. C.
8. The method of claim 1, wherein the first gas stream is a
retentate stream and the second gas stream is a permeate
stream.
9. The method of claim 1, wherein contacting the selectivated
adsorbent with an input gas stream comprises adsorbing, by the
selectivated adsorbent, at least a portion of the second component
during the contacting, the method further comprising desorbing at
least a portion of the adsorbed second component to form a desorbed
second component portion, wherein the second gas stream comprises
at least a portion of the desorbed second component portion.
10. The method of claim 1, further comprising desorbing at least a
portion of the barrier compound during contacting the selectivated
adsorbent or membrane with the input gas stream.
11. The method of claim 1, wherein the minimum dimension for the
barrier compound is from about 4.05 Angstroms to about 5.65
Angstroms.
12. The method of claim 1, wherein the amount of barrier compound
adsorbed by the microporous material is about 20% or less of the
saturation loading (q.sub.s).
13. A method for performing a gas separation in a swing adsorber
unit, comprising: contacting an adsorbent comprising an 8-member
ring zeolite in a swing adsorber unit with a barrier compound under
effective conditions to selectivate the adsorbent, the 8-member
ring zeolite being a DDR type zeolite, ZSM-58, Sigma-1, or a
combination thereof; wherein the minimum molecular dimension of the
barrier compound is at least 0.4 Angstroms greater than the maximum
dimension of the largest hard sphere that can diffuse along any
direction of the adsorbent; wherein the effective conditions for
contacting the adsorbent or membrane with a barrier compound
comprise a temperature from about 50.degree. C. to about
350.degree. C. and a total pressure of about 100 psig (about 690
kPag) to about 2000 psig (about 13.8 MPag), wherein the barrier
compound is present either as a liquid or as a gas with a partial
pressure of the wherein compound that is at least about 10% of the
saturation vapor pressure; wherein selectivation of the adsorbent
or membrane comprises diffusion of barrier compound molecules
through the pores of the 8-member microporous material; contacting
the selectivated adsorbent with an input gas stream containing a
first component and a second component to form an output gas stream
enriched in the first component relative to the input gas stream,
the selectivated adsorbent adsorbing at least a portion of the
second component during the contacting; desorbing at least a
portion of the adsorbed second component to form a desorbed second
component portion; and collecting a gas stream comprising at least
a portion of the desorbed second component portion, the gas stream
being enriched in the second component relative to the input gas
stream.
14. The method of claim 13, wherein the swing adsorber unit is a
pressure swing adsorber unit, a temperature swing adsorber unit, a
rapid cycle pressure swing adsorber unit, or a rapid cycle
temperature swing adsorber unit.
15. The method of claim 13, further comprising repeating said
contacting the selectivated adsorbent, desorbing, and collecting
for a plurality of cycles.
16. The method of claim 15, further comprising desorbing the
barrier compound after the plurality of cycles of repeating said
contacting, desorbing, and collecting.
17. The method of claim 13, wherein a methane diffusivity D.sub.CH4
satisfies the following relationship:
D.sub.CH4<3.times.10.sup.-13*[t.sub.adsorb/(4seconds)].sup.2*[d.sub.cr-
ystal/(15microns)].sup.2{in m2/s}.
18. A method for performing a gas separation in a swing adsorber
unit, comprising: contacting an adsorbent comprising a microporous
material in a swing adsorber unit with a barrier compound under
effective conditions to selectivate the adsorbent, the microporous
material having pores characterized by a first dimension of a
largest hard sphere that can diffuse along any direction in the
pores and the barrier compound having a second dimension
representing a minimum dimension of the compound, wherein the
second dimension is between 10% greater than and 60% greater than
the first dimension; wherein the effective conditions for
contacting the adsorbent or membrane with a barrier compound
comprise a temperature from about 50.degree. C. to about
350.degree. C. and a total pressure of about 100 psig (about 690
kPag) to about 2000 psig (about 13.8 MPag), wherein the barrier
compound is present either as a liquid or as a gas with a partial
pressure of the wherein compound that is at least about 10% of the
saturation vapor pressure; wherein selectivation of the adsorbent
or membrane comprises diffusion of barrier compound molecules
through the pores of the microporous material; contacting the
selectivated adsorbent with an input gas stream containing a first
component and a second component to form an output gas stream
enriched in the first component relative to the input gas stream,
the selectivated adsorbent adsorbing at least a portion of the
second component during the contacting; desorbing at least a
portion of the adsorbed second component to form a desorbed second
component portion; and collecting a gas stream comprising at least
a portion of the desorbed second component portion, the gas stream
being enriched in the second component relative to the input gas
stream.
Description
FIELD OF THE INVENTION
Systems and methods are described for performing gas separation
using adsorbent materials.
BACKGROUND OF THE INVENTION
Removal of contaminants or impurities from a gas phase stream is a
commonly encountered process in petroleum and natural gas
processing. For example, many natural gas streams contain at least
some CO.sub.2 in addition to the desired CH.sub.4. Additionally,
many refinery processes generate a gas phase output that includes a
variety of species, such as CH.sub.4 and CO.sub.2, that are gases
at standard temperature and pressure. Performing a separation on a
gas phase stream containing CH.sub.4 can allow for removal of an
impurity and/or diluent such as CO.sub.2 or N.sub.2 under
controlled conditions. Such an impurity or diluent can then be
directed to other processes, such as being directed to another use
that reduces the loss of greenhouse gases to the environment.
U.S. Patent Application Publication No. 2008/0282885 describes
systems and methods for removing CO.sub.2, N.sub.2, or H.sub.2S
using a swing adsorption process. One type of adsorbent that can be
used in the swing adsorption process is an 8-ring zeolite, such as
a DDR type zeolite.
U.S. Pat. No. 7,255,725 describes a porous inorganic membrane
containing carbon and a process for use of such a membrane. A
porous carbon-free inorganic membrane (such as a zeolite) is
treated with a hydrocarbon-type feed under temperature and time
conditions that are suitable for depositing carbon by chemical
reaction on the inorganic membrane. The carbon-containing membrane
is then maintained at a temperature higher than the deposition
temperature for a period of time prior to performing a membrane
separation. The membrane is described as being useful for
separating non-condensable gases, such as CO.sub.2, CH.sub.4, or
H.sub.2, from a hydrocarbon feed.
International Publication No. WO 2006/017557 describes membranes
for highly selective separations. After calcination, a molecular
sieve membrane such as SAPO-34 is treated with a modifying agent
such as ammonia. Such a treated membrane is described as being
suitable for improving membrane separation of CO.sub.2 from
CH.sub.4 where the amount of CH.sub.4 in a permeate through the
membrane is reduced. Other modifying agents are described as
silanes and/or amines that react with acid sites of zeolites, and
polar compounds such as ethanol.
SUMMARY OF THE INVENTION
In one aspect, a method for performing a gas separation is
provided. The method includes contacting an adsorbent or membrane
comprising an 8-member ring zeolite or other 8-member ring
microporous material with a barrier compound under effective
conditions to selectivate the adsorbent or membrane, the barrier
compound having a minimum dimension of at least about 4.05
Angstroms and a maximum dimension of about 25 Angstroms or less;
contacting the selectivated adsorbent or membrane with an input gas
stream containing a first component and a second component to form
a first gas stream enriched in the first component relative to the
input gas stream; and collecting a second gas stream enriched in
the second component relative to the input gas stream.
In another aspect, a method for performing a gas separation in a
swing adsorber unit is provided. The method includes contacting an
adsorbent comprising an 8-member ring zeolite or other 8-member
ring microporous material in a swing adsorber unit with a barrier
compound under effective conditions to selectivate the adsorbent,
the 8-member ring zeolite being a DDR type zeolite, ZSM-58,
Sigma-1, or a combination thereof; contacting the selectivated
adsorbent with an input gas stream containing a first component and
a second component to form an output gas stream enriched in the
first component relative to the input gas stream, the selectivated
adsorbent adsorbing at least a portion of the second component
during the contacting; desorbing at least a portion of the adsorbed
second component to form a desorbed second component portion; and
collecting a gas stream comprising at least a portion of the
desorbed second component portion, the gas stream being enriched in
the second component relative to the input gas stream.
In yet another aspect, a method for performing a gas separation in
a swing adsorber unit is provided. The method includes contacting
an adsorbent comprising a microporous material in a swing adsorber
unit with a barrier compound under effective conditions to
selectivate the adsorbent, the microporous material having pores
characterized by a first dimension of a largest hard sphere that
can diffuse along any direction in the pores and the barrier
compound having a second dimension representing a minimum dimension
of the compound, wherein the second dimension is between 10%
greater than and 60% greater than the first dimension; contacting
the selectivated adsorbent with an input gas stream containing a
first component and a second component to form an output gas stream
enriched in the first component relative to the input gas stream,
the selectivated adsorbent adsorbing at least a portion of the
second component during the contacting; desorbing at least a
portion of the adsorbed second component to form a desorbed second
component portion; and collecting a gas stream comprising at least
a portion of the desorbed second component portion, the gas stream
being enriched in the second component relative to the input gas
stream.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1 and 2 show zero length chromatography results for diffusion
of gases under various conditions.
FIG. 3 shows CO.sub.2 adsorption isotherms under various
conditions.
FIGS. 4 and 5 show zero length chromatography results for diffusion
of gases under various conditions.
FIG. 6 shows CO.sub.2 adsorption isotherms under various
conditions.
FIG. 7 shows the working capacity for a selectively passivated
adsorbent over time relative to the equilibrium adsorption for a
non-passivated adsorbent.
FIG. 8 shows diffusion coefficients at 30.degree. C. for various
alkanes.
DETAILED DESCRIPTION OF THE EMBODIMENTS
Overview
In various aspects, systems and methods are provided for improving
separation of gas phase streams using an adsorbent, such as a
kinetically selective adsorbent. In the discussion herein,
kinetically selective adsorbents refer to adsorbents with different
diffusion coefficients with regard to the rate of transport of at
least a first compound relative to a at least a second compound. An
example of a suitable adsorbent is an 8-member ring zeolite
adsorbent, such as a DDR type zeolite. Suitable gas phase streams
can include at least one hydrocarbon, such as methane or a
(gaseous) hydrocarbon containing at least one saturated
carbon-carbon bond, and at least one additional component, such as
CO.sub.2 or N.sub.2. Without being bound by any particular theory,
it is believed that the selectivity of the adsorbent can be
improved by the barrier compounds being inserted into (or at least
blocking) the pore structure of an adsorbent material. By entering
(blocking) the pore structure, the barrier compounds can influence
the effective pore mouth area and/or pore volume available for
entry of a potential adsorbate into an adsorbent material. Thus,
even though two potential adsorbates may have sizes suitable for
adsorption into the pore network of an adsorbent, the presence of
the barrier compounds can alter the relative ability of the
different potential adsorbates to enter into and/or move within the
pores of the adsorbent. Optionally, the barrier compounds may also
contribute to formation of a layer on the surface of the adsorbent.
Such an optional layer of barrier compounds may also alter the
ability of potential adsorbates to enter (block) the pores of an
adsorbent material. It is further noted that, while the above
description relates to modifying the kinetics of entry of compounds
into an adsorbate, similar methods can be used to modify the
kinetics of entry of compounds into a membrane, e.g., based on an
8-member ring zeolite structure.
More generally, suitable adsorbents for separation of gas phase
streams can include microporous materials (including zeolites),
such as SAPO materials or other types of structures including
framework atoms other than Si and Al in a zeolite-type framework
structure. Additionally, microporous materials (including zeolites)
with frameworks having other ring sizes may form suitable
adsorbents, such as zeolites or other microporous materials with
10-member ring structures, 12-member ring structures, or structures
with rings of other numbers of atoms.
Preferably, the barrier compounds can alter the kinetics for entry
of a molecular species into the pores of the adsorbent (or
membrane) while having a reduced or minimal impact on the
adsorption (and/or membrane transport) of other targeted molecular
species. This can allow a rate of adsorption for a desired
adsorbate compound to be maintained while improving the selectivity
against adsorption for adsorption of a different compound. In other
embodiments, the kinetics for entry of both a desired adsorbate and
another compound (or compounds) can be altered by the barrier
compounds, with the alteration of the kinetics of the desired
adsorbate being less than the alteration of the kinetics for the
other compound(s), so that the selectivity for preferentially
adsorbing the desired adsorbate compound can advantageously be
improved. For example, CH.sub.4 and C.sub.2H.sub.6 are examples of
compounds where the adsorption kinetics can be impacted by the
presence of certain barrier compounds. N.sub.2 and CO.sub.2 are
examples of compounds that experience a reduced and/or minimized
impact on adsorption kinetics due to the presence of certain
barrier compounds. Thus, N.sub.2 and CO.sub.2 represent potential
desired adsorbate compounds.
In swing adsorption and membrane separation processes,
selectivation of an adsorbent material using barrier compounds can
improve the selectivity of an adsorbent for adsorbing a first
compound preferentially relative to a second compound. For example,
this can allow a product stream of a desired hydrocarbon (such as
CH.sub.4 or C.sub.2H.sub.6) to include a greater yield of the
desired hydrocarbon as well as an increased weight percentage of
the desired hydrocarbon relative to an input gas stream. This can
be achieved by preferentially decreasing the adsorption kinetics of
the desired hydrocarbon relative to the adsorption kinetics of one
or more other components of the input gas stream (such as N.sub.2
or CO.sub.2). Barrier compounds being inserted into (blocking) a
pore network and/or forming a barrier layer for improving
separation can be used, for example, in a swing adsorber apparatus
for performing gas phase separations. Additionally, by slowing the
rate at which a desired hydrocarbon diffuses into the adsorbent,
the cycle time requirements for the adsorbent layer/material can be
relaxed. This can allow for implementation of adsorption and
regeneration cycles that may be more compatible with the timing
required for operation of large scale valves and structured
adsorbent beds. In a membrane separation application, the presence
of a barrier compound can increase the relative likelihood for a
component (such as CO.sub.2 or N.sub.2) to enter and diffuse
through the membrane, as compared with the likelihood for another
component (such as CH.sub.4) to enter and diffuse through the
membrane.
Preferably, the barrier layer does not significantly alter the
adsorption capacity of the adsorbent (or desired permeate
component). In a preferred embodiment, the presence of the barrier
layer only reduces the adsorption capacity of the adsorbent by 40%
or less, e.g., by less than 20% or by less than 10%. Even though
the adsorption capacity can be somewhat reduced by the barrier
layer, the performance of the adsorbent can improve, because of
improvements in the kinetic selectivity.
Preferably, the barrier layer can be formed using a compound that
slowly diffuses through the pore structure of the adsorbent. This
can allow the formation of a relatively stable layer that does not
require frequent regeneration or replenishment. By forming the
barrier layer from a slowly diffusing compound, the layer can be
concentrated in/among pores away from the center of the adsorbent
(i.e., pores near the surface of the adsorbent). Having the barrier
formed in/among pores near/on the surface of the adsorbent can help
to improve the kinetic selectivity by immediately blocking or
retarding the transport, and hence by decreasing the adsorption
kinetics of the species to be kinetically excluded from the
adsorbent. At the temperature of the desired separation process,
the diffusion coefficient of the compound used to form the barrier
layer can preferably be less than 10.sup.-15 m.sup.2/s, e.g., less
than 10.sup.-19 m.sup.2/s or less than 10.sup.-23 m.sup.2/s.
In some preferred embodiments, the adsorbent can preferentially
adsorb a second compound relative to a first compound prior to (or
without) use of barrier compounds. In such embodiments, use of
barrier compounds can optionally further enhance the selectivity of
the adsorbent in membrane or swing adsorption processes. In other
embodiments, barrier compounds can be used for adsorbents that
otherwise have little or no adsorption selectivity for a second
compound relative to a first compound. In such other embodiments,
use of barrier compounds can introduce selective adsorption into an
adsorbent system.
Depending on the embodiment, a blocking compound can be introduced
to an adsorbent to form a barrier layer at various times. One
option can be to introduce the blocking compound after synthesizing
the adsorbent crystals but before incorporating the crystals into a
contactor or bed structure. Another option can be to introduce the
blocking compound after forming an adsorbent bed incorporating the
crystals. Still another option can be to introduce the blocking
compound after an adsorbent bed incorporating the crystals is used
to construct a contactor, such as a swing adsorption vessel.
Removal of CO.sub.2 from a stream containing small hydrocarbons can
be beneficial for a variety of reasons. For example, in a stream
being used as a fuel for combustion, CO.sub.2 can act as an inert
diluent. Too much CO.sub.2 in a fuel stream can inhibit a
combustion reaction. Natural gas for sale as a fuel can often have
a specification on the maximum amount of diluents present. In some
embodiments, the amount of inerts present in a natural gas stream
can preferably be about 2 vol % or less. Additionally, any CO.sub.2
that enters into a combustion reaction can add to the greenhouse
gases generated by the reaction. It can also be beneficial to
remove N.sub.2 from a methane-containing stream (or another stream
containing small hydrocarbons), as N.sub.2 can also act as an inert
diluent during combustion. In addition to acting as a diluent in a
fuel stream, high levels of CO.sub.2 and/or N.sub.2 may increase
the difficulty in liquefying a methane-containing stream, such as a
natural gas stream.
Separation Processes
This invention can be applicable to membrane as well as swing
adsorption processes. Membranes can be formed from adsorbent
materials. For example hydrothermal synthesis processes can produce
zeolite membranes. Zeolites can also be incorporated into mixed
matrix membranes. In a membrane separation process, a flow of a
feed mixture (typically in the gas phase) can be passed over one
side of the membrane. The membrane can selectively transport some
species to the opposite side of the membrane, which either can be
at a lower pressure or can be purged. The species preferentially
transported through the membrane is referred to as the heavy
component, and the species preferentially retained on the feed side
is referred to as the light component (regardless of their relative
molar masses). A transport rate of molecules through the membrane
can be determined by kinetics and equilibrium adsorption. Enhancing
the kinetic selectivity by selectivation can improve the retention,
and hence the recovery, e.g., of the light component.
All swing adsorption processes have an adsorption step in which a
feed mixture (typically in the gas phase) is flowed over an
adsorbent that preferentially adsorbs a more readily adsorbed
component relative to a less readily adsorbed component. A
component may be more readily adsorbed because of kinetic or
equilibrium properties of the adsorbent. The adsorbent can
typically be contained in a contactor that is part of the swing
adsorption unit. The contactor can typically contain an engineered
structured adsorbent bed or a particulate adsorbent bed. The bed
can contain the adsorbent and other materials such as other
adsorbents, mesopore filling materials, and/or inert materials used
to mitigated temperature excursions from the heat of adsorption and
desorption. Other components in the swing adsorption unit can
include, but are not necessarily limited to, valves, piping, tanks,
and other contactors.
The method of adsorbent regeneration designates the type of swing
adsorption process. Pressure swing adsorption (PSA) processes rely
on the fact that gases under pressure tend to be adsorbed within
the pore structure of the microporous adsorbent materials.
Typically, the higher the pressure, the greater the amount of
targeted gas component that will be adsorbed. When the pressure is
reduced, the adsorbed targeted component is typically released, or
desorbed. PSA processes can be used to separate gases of a gas
mixture, because different gases tend to fill the micropore or free
volume of the adsorbent to different extents due to either the
equilibrium or kinetic properties of the adsorbent. Temperature
swing adsorption (TSA) processes also rely on the fact that gases
under pressure tend to be adsorbed within the pore structure of the
microporous adsorbent materials. When the temperature of the
adsorbent is increased, the adsorbed gas is typically released, or
desorbed. By cyclically swinging the temperature of adsorbent beds,
TSA processes can be used to separate gases in a mixture when used
with an adsorbent selective for one or more of the components in a
gas mixture. Partial pressure purge displacement (PPSA) swing
adsorption processes regenerate the adsorbent with a purge. Rapid
cycle (RC) swing adsorption processes complete the adsorption step
of a swing adsorption process in a short amount of time. For
kinetically selective adsorbents, it can be preferable to use a
rapid cycle swing adsorption process. If the cycle time becomes too
long, the kinetic selectivity can be lost. These swing adsorption
protocols can be performed separately or in combinations. Examples
of processes that can be used in combination are RCPSA, RCTSA,
PTSA, and PPTSA. Selectivation can be used to improve the
performance of all swing adsorption processes.
Swing adsorption processes can be applied to remove a variety of
target gases from a wide variety of gas mixtures. The "light
component" as utilized herein is taken to be the species or
molecular component(s) not preferentially taken up by the adsorbent
in the adsorption step of the process. Conversely, the "heavy
component" as utilized herein is taken to be the species or
molecular component(s) preferentially taken up by the adsorbent in
the adsorption step of the process.
The selectivation methods described herein can provide an
improvement for kinetically controlled swing adsorption processes
and, e.g., can increase the recovery of the light component. In
kinetically controlled swing adsorption processes, at least a
portion (and preferably a majority) of the selectivity can be
imparted due to the transport diffusion coefficient in the
micropores and free volume of the adsorbent of the light species
being less than that of the heavier species. Also, in kinetically
controlled swing adsorption processes with microporous adsorbents,
such as 8-member ring zeolite adsorbents, the diffusional
selectivity can arise from diffusion differences in the micropores
of the adsorbent and/or from a selective diffusional surface
resistance in the crystals or particles that make-up the adsorbent.
Kinetically controlled swing adsorption processes are in contrast
to equilibrium controlled swing adsorption processes, where the
equilibrium adsorption properties of the adsorbent tend to control
selectivity. The improvement in the kinetic selectivity can be such
that the total recovery, e.g., of the light component introduced
into the process, achieved in the swing adsorption process can be
greater than about 80 mol %, e.g., greater than about 85 mol %,
greater than about 90 mol %, or greater than about 95 mol %.
Recovery of the light component can be defined as the time averaged
molar flow rate of the light component in the product stream
divided by the time averaged molar flow rate of the light component
in the feedstream. Similarly, recovery of the heavy component can
be defined as the time averaged molar flow rate of the heavy
component in the product stream divided by the time averaged molar
flow rate of the heavy component in the feedstream.
It is-possible to remove two or more contaminants simultaneously,
but, for convenience, the component(s) to be removed by selective
adsorption is(are) typically referred to herein in the singular and
as a contaminant or heavy component.
The selectivation methods described herein can provide an
improvement in the kinetic selectivity of the adsorbent material
that can be translated through proper design into an improvement in
desired component recovery in a kinetically controlled swing
adsorption process and/or membrane separation process.
Adsorbent Contactors and Beds
The term "adsorbent contactor," as utilized herein, includes both
structured and unstructured adsorbent contactors. The adsorbent
contactor is the portion of the swing adsorption unit where the
feed gas is contacted with the adsorbent. In a TSA process, the
contactor may contain a means to heat and cool the adsorbent, such
as heating and cooling channels. Each contactor can contain one or
more adsorbent beds. Beds are sections or portions of the contactor
that contain adsorbent. Each bed can contain a single adsorbent or
a mixture of different adsorbents. All beds in a contactor do not
have to contain the same adsorbent.
In some embodiments, the bed in the contactor can comprise a
packing containing at least solid inert particles and pellets
containing an adsorbent. The inert particles can be incorporated
into the bed to help manage heat of adsorption and desorption. The
pellets containing the adsorbent can typically include adsorbent
particles, pores, and a binder. Pellets can often be formed in
spray drying or extrusion processes. Inert particles can typically
have dimensions ranging from about 100 microns to about 10 cm, but
any suitable particle dimensions can be used, depending on design.
Pellets containing the adsorbent can typically have dimensions
ranging from about 250 microns to about 1 cm, but again any
suitable particle dimensions can be used, depending on design. Mass
transfer can be improved by using smaller sized pellets; however,
pressure drop through the bed can tend to increase with decreasing
size.
One example of an engineered adsorbent contactor is a parallel
channel contactor, which can be suitable for use in a variety of
swing adsorption processes. The bed structure for an adsorbent
contactor composed of parallel channel contactors can include fixed
surfaces on which the adsorbent or other active material is held.
Parallel channel contactors can provide significant benefits over
conventional gas separation methods, such as vessels containing
adsorbent beads or extruded adsorbent particles. "Parallel channel
contactors" are defined herein as a subset of adsorbent contactors
comprising structured (engineered) adsorbents in beds with
substantially parallel flow channels. These flow channels may be
formed by a variety of means. In addition to the adsorbent
material, the bed structure may contain one or more items such as,
but not limited to, support materials, heat sink materials, and
void reduction components.
In a swing adsorption apparatus with parallel contactor channels,
the walls of the channels in the beds can contain the adsorbent,
for example uniform sized 8-ring zeolite crystals. The beds in the
contactor may optionally contain a thermal mass (heat transfer)
material to help control heating and cooling of the adsorbent of
the contactor during both the adsorption and desorption steps of a
pressure swing adsorption process. Heating during adsorption can be
caused by the heat of adsorption of molecules entering the
adsorbent. The optional thermal mass material can also help control
cooling of the contactor during the desorption step. The thermal
mass can be incorporated into the flow channels of the beds in the
contactor, incorporated into the adsorbent itself, and/or
incorporated as part of the wall of the flow channels. When it is
incorporated into the adsorbent, it can be a solid material
distributed throughout the adsorbent layer and/or it can be
included as a layer within the adsorbent. When it is incorporated
as part of the wall of the flow channel, the adsorbent can be
deposited or formed onto the wall. Any suitable material can be
used as the thermal mass material in the practice of the present
invention. Non-limiting examples of such materials include metals,
ceramics, and polymers. Non-limiting examples of preferred metals
include steel, copper, and aluminum alloys. Non-limiting examples
of preferred ceramics include silica, alumina, and zirconia. An
example of a preferred polymer that can be used in the practice of
the present invention is a polyimide.
Depending upon the degree to which the temperature rise is to be
limited during the adsorption step, the amount of thermal mass
material used can range from about 0.1 to about 25 times the mass
of the microporous adsorbent of the contactor, e.g., from about
0.25 to 5 times the mass, from about 0.25 to 2 times the mass, or
from about 0.25 to 1 times the mass. In a preferred embodiment, an
effective amount of thermal mass can be incorporated into the
contactor. The effective amount of thermal mass can be an amount
sufficient to maintain the thermal rise of the adsorbent during the
adsorption step to less than about 100.degree. C., e.g., less than
about 50.degree. C. or less than about 10.degree. C.
Channels in contactors, also sometimes referred to as "flow
channels" or "gas flow channels", are paths in the contactor that
allow gas flow through. Generally, flow channels can provide for
relatively low fluid resistance coupled with relatively high
surface area. Flow channel length can advantageously be sufficient
to provide the mass transfer zone, which length can be at least a
function of the fluid velocity and of the surface area to channel
volume ratio. The channels can be configured to minimize pressure
drop along the length of channels. In many embodiments, a fluid
flow fraction entering a channel at the first end of the contactor
does not communicate with any other fluid fraction entering another
channel at the first end until the fractions recombine after
exiting at the second end. In parallel channel contactors, channel
uniformity can be important in the beds to ensure that
(substantially all of) the channels are being effectively utilized
and that the mass transfer zone is substantially equally contained.
Both productivity and gas purity can suffer if there is excessive
channel inconsistency. If one flow channel is larger than an
adjacent flow channel, premature product break through may occur,
which can lead to a reduction in the purity of the product gas, in
some cases to unacceptable purity levels. Moreover, devices
operating at cycle frequencies greater than about 50 cycles per
minute (cpm) can require greater flow channel uniformity and less
pressure drop than those operating at lower cycles per minute.
Further, if too much pressure drop occurs across the bed, then
higher cycle frequencies, such as on the order of greater than 3
cpm, may not readily be achieved.
The dimensions and geometric shapes of the beds in parallel channel
contactors can include any suitable for use in swing adsorption
process equipment. Non-limiting examples of geometric shapes
include various shaped monoliths having a plurality of
substantially parallel channels extending from one end of the
monolith to the other; a plurality of tubular members; stacked
layers of adsorbent sheets with and without spacers between each
sheet; multi-layered spiral rolls; bundles of hollow fibers; as
well as bundles of substantially parallel solid fibers. The
adsorbent can be coated onto these geometric shapes or the shapes
can, in many instances, be formed directly from the adsorbent
material plus suitable binder. An example of a geometric shape
formed directly from the adsorbent/binder can be extrusion of a
zeolite/polymer composite into a monolith. Another example of a
geometric shape formed directly from the adsorbent can be extruded
or spun hollow fibers made from a zeolite/polymer composite. An
example of a geometric shape coated with the adsorbent can be a
thin flat steel sheet coated with a microporous, low mesopore,
adsorbent film, such as a zeolite film. The directly formed or
coated adsorbent layer can be itself structured into multiple
layers or the same or different adsorbent materials. Multi-layered
adsorbent sheet structures are described, for example, in U.S.
Patent Application Publication No. 2006/0169142, which is
incorporated by reference herein.
The dimensions of the flow channels can be computed from
considerations of pressure drop along the flow channel. It can be
preferred for the flow channels to have a channel gap from about 5
microns to about 1 mm, e.g., from about 50 microns to about 250
microns. As utilized herein, the "channel gap" of a flow channel is
defined as the length of a line across the minimum dimension of the
flow channel as viewed orthogonal to the flow path. For instance,
if the flow channel is circular in cross-section, then the channel
gap is the internal diameter of the circle. However, if the channel
gap is rectangular in cross-section, the flow gap is the distance
of a line perpendicular to and connecting the two longest sides of
the rectangle (i.e., the length of the smallest side of the
rectangle). It should also be noted that the flow channels can be
of any cross-sectional configuration. In some preferred
embodiments, the flow channel cross-sectional configuration can be
circular, rectangular, square, or hexagonal. However, any geometric
cross-sectional configuration may be used, such as but not limited
to, ellipses, ovals, triangles, various polygonal shapes, or even
irregular shapes. In other preferred embodiments, the ratio of the
adsorbent volume to flow channel volume in the adsorbent contactor
can be from about 0.5:1 to about 100:1, e.g., from about 1:1 to
about 50:1.
In some applications, the flow channels can be formed by laminating
adsorbent sheets together. Typically, adsorbent laminate
applications can have flow channel lengths from about 0.5
centimeter to about 10 meters, e.g., from about 10 cm to about 1
meter, and a channel gap of about 50 microns to about 450 microns.
The channels may contain a spacer or a mesh that acts as a spacer.
For laminated adsorbents, spacers can be used, which are structures
or materials that define a separation between adsorbent laminates.
Non-limiting examples of the type of spacers that can be used in
the present invention include those comprised of dimensionally
accurate: plastic, metal, glass, or carbon mesh; plastic film or
metal foil; plastic, metal, glass, ceramic, or carbon fibers and
threads; ceramic pillars; plastic, glass, ceramic, or metal
spheres, or disks; or combinations or composites thereof. Adsorbent
laminates have been used in devices operating at PSA cycle
frequencies up to at least about 150 cpm. The flow channel length
may be correlated with cycle speed. At lower cycle speeds, such as
from about 20 cpm to about 40 cpm, the flow channel length can be
as long as one meter or more, even up to about 10 meters. For cycle
speeds greater than about 40 cpm, the flow channel length can
typically be decreased and may vary, e.g., from about 10 cm to
about 1 meter. Longer flow channel lengths can be used for slower
cycle PSA processes. RCTSA processes tend to be slower than RCPSA
processes, and, as such, longer flow channel lengths can also be
used with TSA processes.
In an embodiment, at least one bed in the contactors within the
swing adsorption unit can contain selectivated adsorbent. In
various embodiments, the majority of the adsorbent contained within
the bed in the contactors can be selectivated.
Gas Feeds and Adsorbent Materials
The separation methods described herein can be used to perform
separations on a variety of gas phase feeds. One example of a gas
phase feed includes a natural gas feed or stream, such as a natural
gas feed produced at a petroleum production site, or a natural gas
feed or stream from a gas field or shale gas formation. Natural gas
feeds typically contain methane, optionally some larger
hydrocarbons such as C.sub.2-C.sub.4 hydrocarbons, CO.sub.2, and
optionally one or more additional components such as N.sub.2,
H.sub.2S, H.sub.2O, and mercaptans. A natural gas feed may also
contain one or more substances introduced as part of the process
for extracting the natural gas at the production site. Non-limiting
examples of such substances can include glycols such as ethylene
glycol, amines such as methyl diethyl amine, dimethyl disulfide,
and combinations thereof.
Improvements in the recovery of the light component created by
selectivation of adsorbents or membranes can be valuable for
processes used to remove impurities from natural gas streams,
particularly high pressure natural gas streams. It can be desirable
to recover the impurities, also referred to as the "heavy
component(s)", and the methane-rich product, also referred to as
the "light component", at as high a pressure as practical for
operability in natural gas processing. Depending on the embodiment,
a swing adsorption process using a selectivated adsorbent can be
used to obtain methane recovery of greater than about 80 mol %,
e.g., greater than about 85 mol %, greater than about 90 mol %, or
greater than about 95 mol %, even when the natural gas is fed at
relatively high inlet pressures, such as greater than about 50 psig
(about 350 kPag), e.g., at least about 150 psig (about 1.0 MPag),
at least about 450 psig (about 3.1 MPag), at least about 600 psig
(about 4.1 MPag), or at least about 1200 psig (about 8.3 MPag). The
composition of natural gas streams directly from an underground
field (raw natural gas) can vary from field to field. In order to
produce a gas that can be introduced into a pipeline for sale to
residential and commercial fuel markets contaminants, such as
N.sub.2, Hg, mercaptans, and acid gases CO.sub.2 and H.sub.2S,
should be removed to acceptable levels. The levels and impurity
types vary from gas field to gas field and, in some cases, can
comprise the majority of molecules in the produced gas. For
example, it is not uncommon for some natural gas fields to contain
from about 0 to about 90 mol % CO.sub.2, more typically from about
10 mol % to about 70 mol % CO.sub.2.
Other examples of suitable gas phase feeds can include a flue gas
and/or a fuel gas from a refinery process. A variety of processes
can generate a flue gas and/or fuel gas including CO.sub.2 and
small hydrocarbons such as CH.sub.4. Depending on the source of the
flue/fuel gas, it/they may also contain H.sub.2S, H.sub.2, N.sub.2,
H.sub.2O, and/or other components that are gas phase at standard
conditions. Components such as CO.sub.2 and N.sub.2 can act as
diluents reducing the value of such flue gas and/or fuel gas
streams.
In order to improve the value of a gas phase stream, a separation
can be performed to generate at least two product streams. A first
product stream corresponding to the light component can be enriched
in a desired product, such as CH.sub.4 and/or other hydrocarbons,
such as other hydrocarbons generally, other hydrocarbons containing
4 or fewer carbon atoms, or other hydrocarbons containing 3 or
fewer carbon atoms. Preferably, the other hydrocarbon can include
at least one saturated carbon-carbon bond. A second product stream
corresponding to the heavy component can be enriched in one or more
rejected components, such as CO.sub.2 and/or N.sub.2.
One method for performing a separation can be to expose an input
stream to an adsorbent material that can preferentially or
selectively adsorb one or more components of a gas phase stream.
Differences in adsorption can be due to either equilibria or
kinetics. Differences in equilibria can be reflected in competitive
adsorption isotherms and/or can be estimated from single component
isotherms. Differences in kinetics can be reflected in diffusion
coefficients. Processes in which a substantial portion of the
selectivity arises from differences in kinetics are typically
referred to as kinetic separations. For kinetic separations, the
time of the adsorption step can preferably be short enough for the
adsorbent not to equilibrate with the feed stream. As an example,
relatively large pore (>5 .ANG. average pore size) cationic
zeolites can have an equilibrium selectivity allowing CO.sub.2 to
be adsorbed in preference to CH.sub.4, while relatively small pore
(<3.8 .ANG. average pore size) cationic zeolites can have a
kinetic selectivity allowing CO.sub.2 to be adsorbed in preference
to CH.sub.4. A contactor made using a zeolite adsorbent can be used
to selectively adsorb CO.sub.2 from an input gas stream containing
CO.sub.2 and CH.sub.4, resulting in an output stream enriched in
CH.sub.4. For a kinetic adsorbent, the time of the adsorption step
can be set by the zeolite crystal size and the CH.sub.4 diffusion
coefficient. Regeneration of such a kinetic adsorbent can be done
with a pressure swing, a temperature swing, a purge, and/or
displacement. Use of a kinetic adsorbent that weakly adsorbs
CO.sub.2 (i.e., relatively flat adsorption isotherm) can facilitate
regeneration. Highly siliceous zeolites (Si/Al ratio
>.about.100) can often have these types of weak isotherms. A
regeneration process can typically generate a stream enriched in
CO.sub.2 and depleted in hydrocarbons such as CH.sub.4.
One consideration in choosing an adsorbent zeolite (or other
adsorbent material) can be selectivity for a desired separation.
Unless otherwise noted, the term "swing adsorption selectivity" as
used herein is based on binary (pairwise) comparison of the molar
concentration of components in a feed stream and the total number
of moles of these components adsorbed by a particular adsorbent
during the adsorption step of a process cycle under the specific
system operating conditions and feed stream composition. This swing
adsorption selectivity definition can be suitable for a process
cycle that is part of a swing adsorption process, such as a type of
pressure and/or temperature swing adsorption. In order to define
the selectivity, uptake values for components in a feed can be
determined. For a feed that contains at least components A and B,
the adsorption uptake values for components A and B can be defined
as: U.sub.A={change in total moles of A in the adsorbent during the
adsorption step of the swing adsorption process}/{molar
concentration of A in the feed}; and U.sub.B={change in total moles
of B in the adsorbent during the adsorption step of the swing
adsorption process}/{molar concentration of B in the feed}, where
U.sub.A represents the adsorption uptake of component A and U.sub.B
represents the adsorption uptake of component B.
For a feed containing component A, component B, and optionally one
or more additional components, an adsorbent that has a greater
"selectivity" for component A than component B can generally have
at the end of the adsorption step of a swing adsorption process
cycle a greater value for U.sub.A than U.sub.B. Thus, the
selectivity can be defined as: Swing Adsorption
Selectivity=U.sub.A/U.sub.B (for U.sub.A>U.sub.B). For
adsorbents that are dominantly kinetically selective, the swing
adsorption selectivity can depend on cycle time. In such cases,
long adsorption, and hence overall, cycle times can decrease the
swing adsorption selectivity.
Based on the above, suitable adsorbents can be operated with cycle
times to have a swing adsorption selectivity of greater than 1 for
a first component (e.g., component A) relative to a second
component (e.g., component B) in a suitable swing adsorption unit.
For example, a suitable adsorbent for separation of CH.sub.4 from
CO.sub.2 can have a swing adsorption selectivity of greater than 1
for adsorption of CO.sub.2 (component A) relative to CH.sub.4
(component B). After selectivation the swing adsorption selectivity
for a first component over a second component can be at least about
5, such as at least about 10. Some selectivations can have a swing
adsorption selectivity for a first component over a second
component of at least 25, e.g., at least 100.
Examples of components can include molecular nitrogen (N.sub.2),
carbon dioxide (CO.sub.2), hydrogen sulfide (H.sub.25), and methane
(CH.sub.4). According to the definitions above, methane represents
a component corresponding to a potential "second component" while
nitrogen, carbon dioxide, or a combination thereof represent
potential choices for the "first component". One option can be to
selectivate an adsorbent (and/or the corresponding suitable swing
adsorption unit) so that the swing adsorption selectivity for
CO.sub.2 over CH.sub.4 is at least 5, e.g., at least 10, at least
25, or at least 100. Additionally or alternately, an adsorbent can
be selectivated so that it has a swing adsorption selectivity for
N.sub.2 over CH.sub.4 of at least 5, e.g., at least 10 or at least
25. Further additionally or alternately, an adsorbent can be
selectivated to have a swing adsorption selectivity for a
combination two or more of the above components relative to
CH.sub.4 of at least 5, e.g., at least 10 or at least 25 (for
example, in such a situation, a first component can be selected
from CO.sub.2, N.sub.2, or H.sub.2S, and the second component can
be CH.sub.4). In all cases, the selectivation can advantageously
increase the swing adsorption selectivity of the adsorbent.
Equilibrium selectivity may also be used as a factor in selecting
an adsorbent. The methods described herein add kinetic selectivity
onto the equilibrium selectivity in a way that can increase the
swing adsorption selectivity of the adsorbent. Equilibrium
selectivity can be characterized based on long time measurements of
transport or based on slow speed cycle performance. For example,
for the adsorption at .about.40.degree. C. of CO.sub.2 using an
8-member ring DDR-type zeolite adsorbent, CO.sub.2 can approach an
equilibrium level of adsorbed molecules on a time scale (order of
magnitude) of about 0.5 seconds to about 10 seconds for .about.10
micron sized crystals. For this order of magnitude description,
approaching an equilibrium level of adsorption is defined as being
within about 5% of the equilibrium adsorption concentration, e.g.,
within about 2%. For CH.sub.4, an equilibrium level of adsorbed
molecules can usually be approached on a time scale on the order of
tens of seconds. In other words, the equilibrium adsorbed
concentration can be approached at a time between about 2 seconds
and about 200 seconds. In a swing adsorption process with a
relatively fast cycle time, such as a rapid cycle pressure and/or
temperature swing adsorption process, the time in the adsorption
step can be comparable to (or possibly shorter than) the time scale
for CH.sub.4 to approach an equilibrium level of adsorbed
molecules. As a result, in a swing adsorption process with a cycle
time on the order of tens of seconds or less, the adsorption of
CO.sub.2 and CH.sub.4 can be influenced by different factors, e.g.,
for a 10 micron sized DDR adsorbent. The CO.sub.2 adsorption can
have characteristics with greater similarity to equilibrium
adsorption, as the time scale for the swing adsorption cycle can be
long relative to the time scale for equilibration of CO.sub.2
adsorption. By contrast, kinetic adsorption factors can have a
greater influence for CH.sub.4 adsorption, as the time scale for
the swing adsorption process can be roughly comparable to the time
scale for CH.sub.4 to reach equilibrium. As a result, selectivation
can be used to increase kinetic selectivity by mitigating CH.sub.4
adsorption and improving process latitude (by increasing allowable
times spent in the adsorption step). N.sub.2 can be similar to
CO.sub.2 in approaching equilibrium adsorption on a faster time
scale than CH.sub.4.
For the removal of CO.sub.2, N.sub.2, and/or H.sub.2S from natural
gas or another gas stream that contains methane and/or other
C.sub.2-C.sub.4 hydrocarbons, examples of suitable adsorbent
materials can include 8-member ring zeolite materials that have a
kinetic selectivity that can be improved with selectivation for
separation of CO.sub.2 and/or N.sub.2 from the feed stream. An
example of a suitable 8-member ring zeolite in this class of
materials that can be used in swing adsorption processes is zeolite
DDR. Other examples of 8-member ring zeolites include Sigma-1 and
ZSM-58, which have isotypic framework structures to DDR. The
8-member ring zeolite materials can have a Si/Al ratio from about
1:1 to about 10000:1, e.g., from about 10:1 to about 5000:1 or from
about 50:1 to about 3000:1.
8-member ring zeolites like DDR can typically have pore channels
with a window (pore) size on the order of 3-4 Angstroms. For
example, the window (pore) size for a DDR type zeolite is about
3.65 Angstroms. Molecules such as CO.sub.2 and/or N.sub.2, with a
relatively linear configuration, can diffuse more rapidly in a pore
with such a window size as compared to bulkier molecules, such as
methane. As another example, the minimum dimension of
methyldiethylamine is slightly less than 5 Angstroms.
In some embodiments, suitable 8-ring zeolite materials can allow
CO.sub.2 to be rapidly transmitted into zeolite crystals while
hindering the transport of methane, so that it is possible to
selectively separate CO.sub.2 from a mixture of CO.sub.2 and
methane. With selectivation the swing adsorption selectivity can
advantageously improve.
The 8-ring zeolites suitable for use herein can allow CO.sub.2 to
access the internal pore structure through 8-ring windows in a
manner such that the ratio of the effective single component
diffusion coefficients of CO.sub.2 and methane (i.e.,
D.sub.CO2/D.sub.CH4) can be greater than 10, e.g., greater than
about 50 or greater than about 100. Another consideration for cycle
performance in natural gas separations can be the absolute methane
diffusivity, methane being the light component in such kinetically
controlled separations. In combination with the length of the
adsorption step in the cyclic swing adsorption process, the
absolute methane diffusivity can significantly impact methane
recovery in kinetically controlled separations of gas mixtures
containing CO.sub.2 and CH.sub.4. As previously mentioned, it can
be advantageous to maximize this recovery. Process modeling has
shown that, for a cycle with a .about.4 second long adsorption step
in a contactor made using .about.15 micron size crystals, the
effective methane diffusivity of crystals at process conditions can
preferably be below about 3.times.10.sup.-13 m.sup.2/s, in order to
avoid adsorbing too much methane and deleteriously impacting
recovery targets. For other cycle times and/or crystals with other
dimensions, the methane diffusivity D.sub.CH4 can preferably be
defined by D.sub.CH4<3.times.10.sup.-13*[t.sub.adsorb/(4
seconds)].sup.2*(d.sub.crystal/(15 microns)].sup.2 {in m.sup.2/s}
where t.sub.adsorb is the time of the adsorption step in seconds
and d.sub.crystal is the characteristic dimension in microns for
transport through the crystal. These conditions for methane
diffusivity cannot always be met with the combination of sizes of
8-ring zeolite crystals that can be synthesized, the diffusivity of
as-synthesized crystals, and the sizes of crystals that can be
practically incorporated into a contactor. Selectivation can
provide the ability to lower the effective methane diffusion
coefficient into the target range for a practical contactor while
preserving and/or increasing the ratio of D.sub.C02/D.sub.CH4.
Considerations for the bounds on the absolute diffusivity of the
light component can be identical for kinetically controlled
separations of other gas mixtures. For the purpose of providing a
testable measurement of performance, effective diffusion
coefficients (for example, those of CO.sub.2 and CH.sub.4) can be
substituted for transport diffusion coefficients measured for a
pure gas in the Henry's law regime of the adsorption isotherm for
the adsorbent. The loading of molecules in the unselectivated
adsorbent (e.g., zeolite) can be low in the Henry's law regime,
and, in this regime, the Fickian and Stephan-Maxwell diffusion
coefficients can be nearly equal. The effective diffusivity of a
porous crystalline material for a particular sorbate can be
conveniently measured in terms of its diffusion time constant,
D/r.sup.2, wherein D is the Fickian diffusion coefficient
(m.sup.2/s) and the value "r" is the radius of the crystallites (m)
characterizing the diffusion distance. In situations where the
crystals are not of uniform size and geometry, "r" represents a
mean radius representative of their corresponding distributions.
One way to measure the time constant and diffusion coefficient can
be from analysis of standard adsorption kinetics (i.e., gravimetric
uptake) using methods described by J. Crank in "The Mathematics of
Diffusion", 2nd Ed., Oxford University Press, Great Britain, 1975.
Another way to measure the time constant and diffusion coefficient
can be from analysis of zero length chromatography data using
methods described by D. M. Ruthven in "Principles of Adsorption and
Adsorption Processes", John Wiley, NY (1984) and by J. Karger and
D. M. Ruthven in "Diffusion in Zeolites and Other Microporous
Solids", John Wiley, NY (1992).
Another way to obtain the diffusion coefficients can be to measure
the time constant for adsorption and desorption in a cyclic single
component pressure swing adsorption unit. In such a unit the
adsorbent is periodically pressurized and depressurized with a pure
single component gas. Each component of interest (for example
CO.sub.2, CH.sub.4, N.sub.2, and He) can be studied separately.
Measurements with He are often used to simplify interpretation of
the data collected. Pressurization can be done by throwing a valve
to connect a pressurized tank to the cell holding the adsorbent and
then closing the valve. Depressurization can be done by throwing
another valve to connect a low pressure tank to the cell holding
the adsorbent and then closing the valve. The moles adsorbed and
desorbed from the sample can be determined from pressure, volume,
and temperature measurements of the tanks and the cell. A
continuous cyclic process can be created by resetting the tank
pressures after the pressurization and depressurization steps. The
time constant for uptake can be readily converted into an effective
diffusion coefficient using a linear driving force (LDF)
approximation. For example, if the adsorbent is treated as an
isotropic medium with a uniform diameter, then the LDF mass
transfer coefficient (s.sup.-1) can be:
.tau.(LDF).apprxeq.16.2*D.sub.effective*(d.sub.crystal/2).sup.2.
This approximation is commonly used to model swing adsorption
processes.
Additionally or alternately, the 8-ring zeolites described above
can be selectivated to increase their kinetic selectivity for
nitrogen separation from methane, or more generally for separation
of nitrogen from C.sub.1-C.sub.4 or C.sub.1-C.sub.3 hydrocarbons in
a gas stream. The improved kinetic selectivity of the suitable
8-ring materials can allow N.sub.2 to be rapidly transmitted into
zeolite crystals while hindering the transport of methane (or other
small hydrocarbons), so that it is possible to selectively separate
N.sub.2 from a mixture of N.sub.2 and methane. For the removal of
N.sub.2, the ratio of effective single component diffusion
coefficients of N.sub.2 and methane (i.e., D.sub.N2/D.sub.CH4) in
the selectivated adsorbent can be greater than 5, e.g., greater
than about 20, greater than about 50, or greater than about 100.
Depending on the embodiment, considerations for the absolute value
of the methane diffusivity can be similar to those discussed
previously.
Selectivation of Adsorbent Materials
Different preparations of adsorbent material can have significantly
different diffusion coefficients. Even with the same preparation
technique diffusion coefficients can in some instances vary from
batch to batch. This has been especially true for zeolites.
Selectivation can provide a method for mitigating variations in the
diffusivity of the light component that could deleteriously affect
recovery of said light component.
Selectivation processes can be applied to a wide variety of
adsorbent materials. The process of selectivation can be
illustrated with zeolite adsorbents, but it should be understood
that microporous materials with other types of framework atoms
and/or other ring structures can also be selectivated. Zeolites are
crystalline materials that are usually grown in a hydrothermal
synthesis process that often uses a structure directing agent.
After recovering the zeolitic crystalline product from the
hydrothermal synthesis mixture, it is often calcined to remove
template and other material that can block access to the pore
structure inside the crystalline particles. The process of
selectivation can be applied after the pore structure of the
adsorbent has been opened. In this specification, zeolite crystals
that have been calcined to remove the structure directing agent are
referred to as "as-synthesized" materials.
Two different classes of techniques have been developed to
selectivate m adsorbents. The first involves longer term exposures
to lower fugacities of the selectivating molecules, while the
second involves exposure at higher fugacities and temperatures for
shorter periods of time. Without being bound by any particular
theory, it is believed that selectivation of an adsorbent can be
achieved by having the molecules enter the adsorbent and occupy
locations within the pores of the adsorbent. Selectivating
molecules can be concentrated in an absorption front that passes
inward from the exterior surface of the adsorbent. The
concentration of selectivating molecules can decrease dramatically
at distances beyond which the front has penetrated. The
selectivating molecules can act as barrier compounds altering the
ability of potential adsorbates to enter the adsorbent and move
within the adsorbent. This can result in an increase in kinetic
selectivity for the adsorbent. For example, the selectivation by
the barrier compounds can cause the kinetic selectivity for an
8-ring zeolite to increase for gas separation of CO.sub.2 and/or
N.sub.2 from CH.sub.4 or another hydrocarbon. Optionally, the
barrier compounds can also form a passivation layer at the surface
of the adsorbent that improves the kinetic selectivity of the
adsorbent.
Preferably, the barrier compounds can slowly diffuse (diffusion
coefficients less than 10.sup.-15 m.sup.2/s) through the pore
structure of the adsorbent under operating conditions. Very slow
diffusing molecules (diffusion rates less than 10.sup.-19
m.sup.2/s) can be concentrated in pores away from the center of the
adsorbent (i.e., pores near the surface of the adsorbent). In the
extreme limit of ultra-slow diffusion (less than 10.sup.-23
m.sup.2/s), the molecules can be located in a distinct layer that
extends inward from the surface of the adsorbent. Having the
barrier formed in/among pores near/on the surface of the adsorbent
can help to improve the kinetic selectivity in a swing adsorption
process by immediately blocking or retarding the transport and
hence decreasing the adsorption kinetics of the species to be
kinetically excluded from the adsorbent. This can help increase
recovery by not picking-up molecular species to be recovered as a
product during the adsorption step. For 8-ring zeolites, examples
of molecules in the range from slowly diffusing to very slowly
diffusing molecules can include linear alkanes with more than 4
carbons in the chain and primary linear alcohols with more than 3
carbons. For 8-ring zeolites, examples of molecules that can
include very slowly diffusing to ultra-slowly diffusing can include
mono-branched alkanes, di-branched alkanes, 2-octanol, secondary
alcohols, multiple-branched alkanes, branched primary alcohols, and
combinations thereof. Ultra-slowly diffusing molecules can include
tertiary branched paraffins, such as 2,2,4 trimethyl pentane.
Slowly diffusing and ultra-slow diffusing molecules can allow the
formation of a relatively stable selectivation layer that does not
require frequent regeneration or replenishment.
To form an effective barrier that blocks or retards transport, the
maximum loading of the blocking compound in/among the pores near
the surface of the adsorbent should be at least 10% of its
saturation loading (q.sub.s, typically expressed in mmol/gram). In
other embodiments, the maximum loading of the blocking compound
in/among the pores near the surface of the adsorbent can be at
least 40% of its saturation loading, e.g., at least 75% of its
saturation loading. In other embodiments, the loading can be
limited relative to the saturation loading in order to provide at
least some open channel. In such embodiments, the loading can be
20% of the saturation loading or less, e.g., 15% or less. The
region near the surface can be taken to mean a distance from the
surface within 5% of the adsorbent diameter or average particle
size. This maximum loading can be estimated from the fugacity of
the blocking compound in the vapor to which the adsorbent is
exposed. The estimation uses a Langmuir isotherm to approximate the
equilibrium uptake of the blocking compound. The mathematical form
of a Langmuir isotherm can be: q/q.sub.s=b*f*(1+b*f) where q is the
loading (mmol/gram) of the blocking compound, q.sub.s is the
saturation loading (mmol/gram) of the blocking compound, f is the
fugacity (bar) of the blocking compound, and b is the Langmuir
coefficient (1/bar). This condition can be achieved by exposure to
vapor pressures of the barrier compounds with molecular weight of
at least about 50 g/mol, e.g., at least about 60 g/mol, at partial
pressures greater than about 10% of the saturated vapor pressure.
Higher maximum loadings can be achieved by exposure to higher
partial pressures of the barrier compounds, such as at least about
25% of the saturated vapor pressure, at least about 50%, or at
least about 90%. For exposure at very high partial pressures of the
barrier compounds, a liquid phase can form, and suitably high
loadings of the barrier compounds can be achieved. Examples of
suitable barrier compounds can include, but are not limited to,
alkanes (e.g., paraffins), other hydrocarbons, alcohols, other
oxygenates, amines, and sulfur bearing compounds. Paraffinic
species can include linear paraffins, mono-branched paraffins,
multiply-branched paraffins, tertiary branched paraffins, and
combinations thereof. Alcohols can include primary and secondary
alcohols. Oxygenates can include glycols, such as ethylene glycol
and triethylene glycol. Examples of amines can include
methyldiethylamine and dimethylethylamine. One example of a sulfur
bearing compound is dimethyl disulfide. Selectivating with a
blocking compound having a "b" value of greater than 10/bar at the
characteristic temperature of the swing adsorption process can help
to mitigate loss of the blocking compound after it has been loaded
into the adsorbent. Use of blocking compounds with higher b values
(such as at least 100/bar or at least 1,000/bar) at the
characteristic temperature of the swing adsorption process
temperature) can further mitigate loss of the blocking compound
during operation of a swing adsorption or membrane separation
process.
Molecular size can affect both the adsorption strength (b value) as
well as the diffusivity of the blocking compound. Molecular sizes
can be characterized by a maximum and minimum dimension of the
blocking molecule. For example, for a linear paraffin, the maximum
dimension can be set by the stretched out chain length and the
minimum dimension can be set by the chain diameter. Increasing
maximum dimension of a blocking molecule can tend to increase the b
value and lower the diffusivity. When the maximum dimension of the
blocking molecule becomes too big, it cannot selectively block
transport of one molecular species as effectively while allowing
another species to rapidly transport. For 8-member ring zeolite
adsorbents, the maximum dimension of the blocking compound can be
less than about 25 Angstroms and greater than about 4 Angstroms.
Enhanced performance can be achieved for 8-member ring zeolite
adsorbents when the maximum dimension is less than about 15
angstroms and greater than about 6 angstroms. Molecules with
smaller maximum dimensions, such as CO.sub.2, H.sub.2S, methane,
ethane, and H.sub.2O, do not appear to improve the kinetic
selectivity of adsorbents. The minimum dimension of the blocking
molecule can determine its ability to diffuse into a zeolite. As
the minimum dimension of the blocking molecule increases, the
diffusion coefficient can tend to decrease. When the minimum
dimension becomes too large, the diffusion coefficient can fall to
the point that a compound cannot pass into the interior of the
zeolite in a practical amount of time. The size of a molecule that
can pass into the pore structure of a zeolite may be larger than
what might be inferred from the database of zeolite structures
published by the International Zeolite Association (IZA);
http://izasc.ethz.ch/fmi/xsl/IZA-SC/ft.xsl. For each framework
type, the database gives the dimension of a hard sphere that can
diffuse along each direction in a rigid zeolite framework. Because
of the flexibility of the zeolite framework, molecules with minimum
dimensions significantly larger than those inferred from the IZA
database can penetrate into the zeolite pore structure. To be able
to load blocking molecules into the zeolite the minimum molecular
dimension can be less than 2.0 Angstroms greater than the IZA's
dimension of the largest hard sphere that can diffuse along any
direction in a rigid zeolite framework. To be able to have a more
facile loading of blocking molecules into the zeolite their minimum
molecular dimension can be less than 1.5 Angstroms greater than the
IZA's dimension of the largest hard sphere that can diffuse along
any direction in a rigid zeolite framework. To have a diffusion
coefficient small enough to trap the blocking molecules in the
zeolite, the minimum molecular dimension of blocking molecules can
be greater than about 0.4 Angstroms greater than the IZA's
dimension of the largest hard sphere that can diffuse along any
direction in a rigid zeolite framework. Trapping of blocking
molecules can be enhanced if their minimum molecular dimension is
greater than about 0.6 Angstroms greater than the IZA's dimension
of the largest hard sphere that can diffuse along any direction in
a rigid zeolite framework. Additionally or alternately, the
dimensional differences can be expressed as percentages of the
dimension of the largest hard sphere from the IZA database. In such
embodiments, the minimum molecular dimension can be less than about
60% greater than the IZA's dimension of the largest hard sphere
that can diffuse along any direction in a rigid zeolite framework,
e.g., less than about 55% greater, less than about 50% greater,
less than about 45% greater, or less than about 40% greater.
Additionally or alternately in such embodiments, the minimum
molecular dimension can be at least about 10% greater than the
IZA's dimension of the largest hard sphere that can diffuse along
any direction in a rigid zeolite framework, e.g., at least about
15% greater, at least about 20% greater, or at least about 25%
greater.
For example, the IZA database lists the size of the largest hard
sphere that can diffuse along any direction in a rigid DDR zeolite
framework as 3.65 Angstroms. Thus, the preferred minimum dimension
of a suitable blocking molecule for DDR can be between about 4.05
and 5.65 Angstroms. To facilitate loading of the blocking molecules
into DDR, it can be preferred that the minimum dimension of the
blocking molecule be less than about 5.15 Angstroms. To enhance the
trapping (i.e., long term stability) of the blocking compound in
DDR, it can preferred that the minimum dimension of the blocking
molecule be greater than 4.25 Angstroms. For selectivation of a DDR
type zeolite, examples of molecules that fall in these size ranges
can include, but are not limited to, n-hexane, 2-methylpentane,
2-methylhexane, hexanol, 2-hexanol, 2-heptanol, 2-octanol,
2-methyl-1pentanol, 2-methyl-1-hexanol. For a DDR type zeolite,
examples of molecules that fall outside this size range include
toluene, cyclohexyl amine, butylamine, and n-methylpyrrolidone.
Experiments have shown that long term exposures at temperatures
near 100.degree. C. to saturated vapors of these molecules that
fall outside of this size range do not appear to significantly
improve the kinetic selectivity.
Some molecules used to selectivate can react with defects in
zeolite framework structures (such as hydroxyls), chemically
bonding them inside the framework. This can enhance the stability
of the selectivation. For instance, primary and secondary alcohols
can diffuse into the zeolite structure and subsequently react with
hydroxyls and other defects in the structures. This reaction can
trap the molecules so as to be able to resist desorption upon
heating to temperatures of .about.350.degree. C. or higher in an
inert atmosphere (e.g., N.sub.2).
A variety of methods can be used for applying a barrier compound to
an adsorbent. One option can be to expose an adsorbent to the
barrier compound prior to incorporating the adsorbent into a bed of
a swing adsorption unit. In that situation, adsorbent particles can
be treated before formulating them into a bed. Formulation
techniques can include forming a pellet with the selectivated
adsorbent and a binder, casting the selectivated adsorbent and a
binder into a film, wash coating the selectivated adsorbent and
optionally a binder onto a support such as a monolith. In one
embodiment the adsorbent can be selectivated for a period of about
1 hour to about 150 hours, at pressures ranging from about 100 psig
(about 690 kPag) to about 2000 psig (about 13.8 MPag), at
temperatures ranging from about 155.degree. C. to about 350.degree.
C., by direct exposure to a high concentration (>90 mol %) of
the molecules chosen to alter the kinetic selectivity of the
adsorbent. Lower temperatures can slow diffusion of the molecules
chosen for selectivation into the adsorbent, while temperatures
that are too high can lead to thermal degradation of the barrier
compound. Treatment times can be reduced without thermally
degrading the selectivating molecules, e.g., by using temperatures
from about 250.degree. C. to about 310.degree. C. To increase the
degree of loading, it can be desirable to use as high a fugacity as
possible for the treatment. If the selectivating molecules can be
condensed into a liquid phase, then the fugacity usually does not
rise very rapidly with increasing pressure, and the treatment
pressure can be reduced to a point just above (>1 psi above) the
pressure at which the condensation occurs. If there is no liquid
phase transition, then it can be desirable to use a pressure such
that the fugacity (f) is such that the product (b*f) is greater
than 10 at the temperature of the selectivation. This condition can
routinely be met when the selectivation is done at pressures
greater than about 500 psig (about 3.5 MPag). Often, it can be
desirable to use a neat (pure) compound for selectivation. When
this is done, it can be preferred that the purity of the compound
is greater than about 95 mol %, often greater than about 99 mol %.
It can be preferable for the impurities present not to interfere
with the loading and selectivating properties of the compound. When
a mixture of compounds is used for selectivation, it can be
preferred that more than about 95 mol % of the molecules in the
mixture are those targeted for loading into the adsorbent, and it
can be desirable for the purity of the molecules in the mixture to
be greater than about 99 mol %. When the selectivation occurs at
elevated pressure and temperature, the adsorbent can be loaded with
a blocking compound inside an autoclave. When an autoclave (or
batch treater) is used, it can take time for the autoclave to heat
and cool. To provide process latitude without sacrificing
throughput, it can be desirable to have the treatment time be in a
range from about 2 hours to about 20 hours. If a continuous high
pressure/high temperature treater is used (for example, a high
pressure fluid bed treater), shorter average residence times can be
achieved. After the barrier compound is loaded in the adsorbent,
the adsorbent can be cooled for further processing that can
ultimately incorporate it into a bed in a contactor. Because the
barrier compound is believed to be at least partially within the
pore structure of the adsorbent, the barrier compounds can be
maintained within/among the adsorbent during multiple swing
adsorption cycles. Improvement in kinetic selectivity and reduction
in the absolute diffusivity of the light component can vary,
depending on the exact choice of the selectivating molecule as well
as the combination of time, temperature, and pressure conditions
chosen in the selectivation process. It can thus be desirable to
optimize the choice of conditions by measurement of the diffusivity
of the light component. It can be desirable to accurately measure
this quantity, because the recovery of the light component can
depend on the absolute value of its effective diffusivity. Because
the heavy component tends to diffuse rapidly, measurement of its
diffusivity only has to be accurate enough to establish sufficient
kinetic selectivity for the swing adsorption process.
Alternatively, before incorporating an adsorbent structure into a
bed, the adsorbent can be selectivated by exposure to the barrier
compound for times longer than 150 hours. Because of the lower
throughput in manufacturing, this is usually a less desirable
process to make a selectivated adsorbent. With longer exposure
times, selectivation may be done at lower temperatures and/or lower
pressures.
Another option can be to expose the adsorbent to the barrier
compound after it has been incorporated into a bed or a component
to be assembled into a bed to be used in the contactor within the a
swing adsorption unit. When the bed is formed with pellets
containing the adsorbent, the pellets can be selectivated after
they are formed using the procedure that was previously described
to selectivate adsorbent particles. When the beds are designed for
use in a parallel channel contactor, beds can be selectivated
individually using the procedures described to selectivate
adsorbent particles. For example, a bed in a parallel channel
contactor may be a monolith that has been wash coated with the
adsorbent. In such a situation, the monolith can be selectivated in
an autoclave using the procedure described to selectivate adsorbent
particles. In all such situations, the bed can be selectivated
before being incorporated into a swing adsorption unit.
To selectivate beds containing DDR adsorbents, examples of suitable
barrier compounds that can be deposited using the procedures
described can include, but are not limited to, n-hexane,
2-methylpentane, 2-methylhexane, hexanol, 2-hexanol, 2-heptanol,
2-octanol, 2-methyl-1-pentanol, 2-methyl-1-hexanol, and
combinations thereof.
Another option can be to expose the adsorbent to the barrier
compound after beds have been assembled to form a contactor. In
this case, it can be desirable to to be able to heat the contactor
to temperatures greater than 100.degree. C., e.g., greater than
150.degree. C., and/or to expose the contactor to barrier
compound(s) with a molecular weight of at least about 50 g/mol,
such as at least about 60 g/mol, at higher partial pressures, such
as a partial pressure corresponding to at least about 25%, e.g., at
least about 50% or at least about 90%, of thee saturated vapor
pressure of the barrier compound(s). If the temperature is high
enough so that no liquid phase of the barrier compound(s) is
present, then it can be desirable to conduct the exposure at
pressures greater than about 100 psig (about 690 kPag) with at
least about 10 mol % (e.g., at least 50 mol %) of the barrier
compound(s) in the stream. Exposure times can range from hours to
weeks.
Another option can be to perform the selectivation after the
contactors have been installed within a swing adsorption unit. In
this case, the selectivation can be performed when the contactor is
within a functional swing adsorption unit, with a barrier compound
with molecular weight of at least about 50 g/mol, such as at least
about 60 g/mol, being incorporated into the feed gas at partial
pressures greater than about 10%, e.g., at least about 25%, at
least about 50%, or at least about 90%, of the saturated vapor
pressure. Selecting suitable barrier compound concentrations can
depend on the choice of barrier compound, cycle length, temperature
at which the separation process is conducted, and the length of
treatment time, inter alia. Treatment times can be from days to
months. Methods by which barrier compounds can be incorporated into
the feed gas stream can include direct injection and/or use of
vaporizers such as bubblers. It can be preferred to add barrier
compounds to the feed stream, even though, in some instances,
barrier compounds may already be present. As the adsorbent
selectivates due to exposure to the barrier compounds, the
performance of the swing adsorption unit can advantageously change.
It can be desirable to control and plan for the rate of performance
change. Once the performance has reached a desired target, the
concentration of barrier compound in the feed stream can be
reduced. In some instances a small maintenance level of the barrier
compound may be added to the feed to improve the long-term
stability of the selectivation. The goal of the selected conditions
can be to allow the barrier compound to enter the pore network to
improve the kinetic selectivity of the adsorbent, thus lowering the
diffusivity of the light component. It can also desirable for the
selectivation to be maintained between cycles in a manner that does
not cause a large excess of the barrier compound to build up on the
adsorbent surface.
Another option for selectivation can be to include a barrier
compound in a separate gas stream passing through a contactor after
it is installed within a working swing adsorption unit. This stream
can be different from the feed stream, in which case the unit does
not have to be operated in a manner to produce a separation of the
stream. As such, the valve sequencing in the unit need not be the
same as that used for the cyclic swing adsorption process. Streams
containing the barrier compound can be recycled and sent back
through the unit or can be used on a once through basis. When the
stream is recycled, the barrier compound can be replenished either
by direct injection and use of vaporizers such as bubblers. Streams
containing the barrier compound can be sent through the unit at
temperatures greater than those used in the swing adsorption
process. Similarly, the unit can be heated while the selectivating
stream is being sent through the unit. In either case, it can be
desirable to perform the exposure to selectivating molecular
species at temperatures greater than 100.degree. C., e.g., greater
than 150.degree. C. Concentrations of the barrier compound can be
higher than in the case in which it is incorporated into the feed
of a unit performing a separation. In such a case, it can be
preferred that the exposure be done with higher partial pressures
of the barrier compounds, such as at least about 25%, e.g., at
least about 50% or at least about 90%, of the saturated vapor
pressure. If the temperature is high enough so that there is no
liquid phase of the barrier compound present, then it can be
desirable to conduct the exposure at pressures greater than about
100 psig (about 690 kPag) with at least 10 mol % (e.g., at least 50
mol %) of the barrier compound in the stream. Exposure times can be
from hours to weeks.
Any two or more of the options can alternately be combined to
provide for improved selectivity, as desired.
Other Embodiments
Additionally or alternately, the present invention can include one
or more of the following embodiments.
Embodiment 1
A method for performing a gas separation, comprising: contacting an
adsorbent or membrane comprising an 8-member ring zeolite or an
8-member ring microporous material with a barrier compound under
effective conditions to selectivate the adsorbent or membrane,
wherein the barrier compound has a minimum dimension of about 4.05
Angstroms to about 5.65 Angstroms and a maximum dimension of about
25 Angstroms or less for an 8-member ring zeolite or an 8-member
ring microporous material; contacting the selectivated adsorbent or
membrane with an input gas stream containing a first component and
a second component to form a first gas stream enriched in the first
component relative to the input gas stream; and collecting a second
gas stream enriched in the second component relative to the input
gas stream.
Embodiment 2
The method of Embodiment 1, wherein the 8-member ring zeolite is a
DDR type zeolite, Sigma-1, ZSM-58, or a combination thereof.
Embodiment 3
The method of any of the above embodiments, wherein the barrier
compound is a glycol, an amine, an alcohol, an alkane, a sulfur
bearing compound, or a combination thereof, the barrier compound
having a molecular weight of at least 50 g/mol.
Embodiment 4
The method of any of the above embodiments, wherein the barrier
compound is ethylene glycol, triethylene glycol, methyl diethyl
amine, dimethyl ethyl amine, dimethyl disilane, n-hexane,
2-octanol, or a combination thereof.
Embodiment 5
The method of claim 1, wherein the first component comprises or is
CH.sub.4.
Embodiment 6
The method of any of the above embodiments, wherein the second
component comprises or is CO.sub.2, N.sub.2, or a combination
thereof.
Embodiment 7
The method of claim 1, wherein the effective conditions for
contacting the adsorbent or membrane with a barrier compound
comprise a temperature from about 50.degree. C. to about
350.degree. C. and a total pressure of about 100 psig (about 690
kPag) to about 2000 psig (about 13.8 MPag), wherein the blocking
compound is present either as a liquid or as a gas with a partial
pressure of the blocking compound that is at least about 10% of the
saturation vapor pressure.
Embodiment 8
The method of Embodiment 7, wherein the effective conditions for
contacting the adsorbent or membrane with a barrier compound
comprise a temperature of at least about 150.degree. C., such as at
least about 250.degree. C.
Embodiment 9
The method of any of the above embodiments, wherein contacting the
selectivated adsorbent with an input gas stream comprises
adsorbing, by the selectivated adsorbent, at least a portion of the
second component during the contacting, the method further
comprising desorbing at least a portion of the adsorbed second
component to form a desorbed second component portion, wherein the
second gas stream comprises at least a portion of the desorbed
second component portion.
Embodiment 10
The method of any of the above embodiments, further comprising
desorbing at least a portion of the barrier compound during
contacting the selectivated adsorbent or membrane with the input
gas stream.
Embodiment 11
The method of any of the above embodiments, wherein the adsorbent
is part of a swing adsorber unit, the swing adsorber unit being a
pressure swing adsorber unit or a temperature swing adsorber
unit.
Embodiment 12
The method of embodiment 11, wherein the swing adsorber unit is a
rapid cycle pressure swing adsorber unit or a rapid cycle
temperature swing adsorber unit.
Embodiment 13
The method of embodiment 11 or 12, further comprising repeating
said contacting, desorbing, and collecting for a plurality of
cycles.
Embodiment 14
The method of any of embodiments 1-8 or 10, wherein the first gas
stream is a retentate stream and the second gas stream is a
permeate stream.
Embodiment 15
The method of any of the above embodiments, wherein the amount of
barrier compound adsorbed by the microporous material is about 20%
or less of the saturation loading (q.sub.s).
Embodiment 16
A method for performing a gas separation in a swing adsorber unit,
comprising: contacting an adsorbent comprising a microporous
material in a swing adsorber unit with a barrier compound under
effective conditions to selectivate the adsorbent, the microporous
material having pores characterized by a first dimension of a
largest hard sphere that can diffuse along any direction in the
pores and the barrier compound having a second dimension
representing a minimum dimension of the compound, wherein the
second dimension is between 10% greater than and 60% greater than
the first dimension; contacting the selectivated adsorbent with an
input gas stream containing a first component and a second
component to form an output gas stream enriched in the first
component relative to the input gas stream, the selectivated
adsorbent adsorbing at least a portion of the second component
during the contacting; desorbing at least a portion of the adsorbed
second component to form a desorbed second component portion; and
collecting a gas stream comprising at least a portion of the
desorbed second component portion, the gas stream being enriched in
the second component relative to the input gas stream, and
optionally including the subject matter of any one or more of
embodiments 3-15.
EXAMPLES
Example 1
Selectivation of DDR Crystals by Longer Term Exposure to Vapors of
Barrier Compounds at Temperatures Below 150.degree. C.
Selectivation of an adsorbent material can inhibit adsorption of a
lighter component, such as CH.sub.4, to a greater degree than a
heavier component, such as CO.sub.2 and/or N.sub.2. Reduction of
the value of the diffusion coefficient of the lighter component can
play a significant role in improving recovery in membrane and swing
adsorption processes. Thus, even though uptake of both lighter and
heavier components may be reduced, it can be especially
advantageous for the relative uptake rate of CH.sub.4 (or another
lighter component) to be reduced more than the heavier
component.
The difference between CH.sub.4 adsorption and CO.sub.2 and/or
N.sub.2 adsorption on 8-member ring zeolite adsorbents can be seen
in the order of magnitude (or greater) difference in the diffusion
rate of the various components. It has been observed that different
preparations of the 8-member ring zeolite DDR can have
significantly different diffusion coefficients for CH.sub.4, while
maintaining a large ratio between diffusion coefficients for
CO.sub.2 and CH.sub.4. This Example discusses transport and
selectivation that are characteristic for three different batches
of DDR crystals with high Si/Al ratios (>.about.100) that were
found to have similar transport diffusion coefficients. Each
crystal batch appeared to have a relatively uniform particle size.
Characteristic dimension of the batch with the smallest sized
crystals was .about.2 microns, and the characteristic dimension of
the batch with the largest size crystals was .about.30 microns. The
majority of the selectivation results presented come from a batch
with a .about.16 micron characteristic crystallite (particle) size.
Selectivation characteristics discussed herein are believed to also
apply to other preparations of DDR materials with different
CH.sub.4 diffusion coefficients in the as-synthesized crystals. The
details of this Example are also believed to be applicable to the
selectivation of other zeolites.
In this example, for DDR type 8-member ring zeolites, the transport
diffusion coefficients for CH.sub.4 at temperatures between about
50.degree. C. and about 100.degree. C. were found to range from
about 1.times.10.sup.-8 cm.sup.2/s to less than 1.times.10.sup.-9
cm.sup.2/s. By contrast, in this same temperature range, the
transport diffusion coefficients for CO.sub.2 were found to range
from about 1.times.10.sup.-6 cm.sup.2/s to about 1.times.10.sup.-7
cm.sup.2/s. The transport diffusion coefficient for N.sub.2 was
similar in order of magnitude to the transport diffusion
coefficient for CO.sub.2, but had less temperature variance. As a
result, diffusion of CO.sub.2 and/or N.sub.2 in an 8-ring zeolite
adsorbent appeared to occur on a rapid time scale relative to the
much slower diffusion of CH.sub.4. Due to the faster diffusion
rates, the adsorption of CO.sub.2 and/or N.sub.2 can be influenced
to a greater degree by equilibrium diffusion properties, while
adsorption of CH.sub.4 can be influenced to a greater degree by
kinetic diffusion properties. Thus, a method inhibiting kinetic
diffusion can likely have a greater impact on CH.sub.4 adsorption
than CO.sub.2 and/or N.sub.2 adsorption.
One technique for investigating the adsorption properties of an
adsorbent is zero length chromatography. Zero length chromatography
(ZLC) determines diffusion coefficients from measurements of rates
at which adsorbed molecules are purged from samples after rapidly
switching from adsorption to desorption conditions. Analysis of
zero length chromatography data methods is described, e.g., by D.
M. Ruthven in "Principles of Adsorption and Adsorption Processes",
John Wiley, NY (1984) and by J. Karger and D. M. Ruthven in
"Diffusion in Zeolites and Other Microporous Solids", John Wiley,
NY (1992). An example of pure component zero length chromatography
data on uniform sized DDR crystals is shown in FIG. 1. The ZLC
apparatus used in this experiment was modified from that
traditionally used in the literature to accommodate and operate
with a larger amount of sample than the literature would
indicate.
Data shown in FIG. 1 are for ethane diffusion out of DDR crystals
at .about.50.degree. C. into a He stream as the flow through the
ZLC cell is varied. The DDR crystals had an average size of about
16 .mu.m. The flow through the ZLC was varied from .about.5.2
mL/min to .about.26 mL/min, as shown in the figure. In FIG. 1, the
vertical axis shows a log scale of the concentration of ethane
diffusing out of the DDR crystals relative to the amount of ethane
diffusion at time 0. Thus, FIG. 1 shows the change in the amount of
diffusion out of the crystals as a function of time. For each curve
in FIG. 1, both a measured value and fits of the values to a model
are shown.
As shown in FIG. 1, models using a single diffusion process
throughout the crystal appear to match ZLC data sets taken for
CH.sub.4, CO.sub.2, and C.sub.2H.sub.6 diffusion in DDR. For a
known uniform crystal size, the model fits involved two parameters.
One parameter was the diffusion coefficient while a second
parameter was Henry's constant for molecular adsorption. In the
fits to the model, the Henry's constant matched one determined
experimentally by independent equilibrium adsorption measurements.
The value fits to the data also matched the Henry's constant
generated from theoretical predictions. The model fits to CO.sub.2
and CH.sub.4 data indicated that, for DDR crystals, mass transfer
was controlled primarily by diffusion through the volume of the
crystal.
ZLC studies can also be used to investigate diffusion out of
crystals that have been exposed to one or more barrier compounds
and/or foulants. As an initial test, DDR crystals were exposed to a
base gas mixture containing CO.sub.2, H.sub.2S, CH.sub.4, and
optionally small amounts of C.sub.2-C.sub.2 components for a
.about.1 month exposure period. The DDR crystals were exposed to
the base gas mixture at a pressure of about 850 psig (about 5.9
MPag) and a temperature of .about.100.degree. C. FIG. 2 shows a ZLC
data set for methane diffusion out of DDR crystals after exposure
to base fouling. The model fits in FIG. 2 used parameters similar
to the parameters for fresh crystals. As shown in FIG. 2, exposure
to the base gas resulted in little or no change of measured methane
diffusion relative to a single diffusion process model.
Exposure to base gas also appeared to have minimal or no impact on
CO.sub.2 adsorption. FIG. 3 shows an example of CO.sub.2 adsorption
isotherms for DDR crystals in a fresh state and after exposure to
several types of base gas. Due to the faster diffusion rate of
CO.sub.2 in DDR, an adsorption isotherm provided a clearer
indication of differences in adsorption activity. As shown in FIG.
3, the adsorption isotherms for CO.sub.2 on DDR crystals exposed to
various types of base gas were comparable to the adsorption
isotherm for fresh crystals.
The data in FIGS. 2 and 3 indicated that exposure to CO.sub.2,
H.sub.2S, and H.sub.2O did not appear to have a significant impact
on the adsorption characteristics of the DDR crystals. This finding
was in contrast to barrier compounds according to the invention,
which appeared to modify the transport characteristics of such
crystals.
Additional ZLC studies of CH.sub.4 transport in DDR crystals were
performed where an additional single barrier compound and/or
foulant was included in the base gas mixture. The exposure
conditions were otherwise similar to the base gas exposure
conditions mentioned above. CO.sub.2 adsorption isotherms were also
generated for DDR crystals exposed to base gases including the
various barrier compounds and/or foulants at or near their
saturated vapor pressure in the base gas mixture.
In addition to the above ZLC studies and CO.sub.2 adsorption
isotherms, alteration of both CH.sub.4 and CO.sub.2 adsorption was
shown for several barrier compounds. The barrier compounds that
impacted the adsorption characteristics included ethylene glycol,
triethylene glycol, dimethyl disulfide (DMDS), hexane, methyl
diethyl amine (MDEA), and dimethyl ethyl amine (DMEA). By contrast,
additional ZLC studies and CO.sub.2 adsorption isotherms showed
little or no impact on the adsorption properties of DDR for several
foulants with a characteristic (minimum) dimension greater than the
barrier compounds. For example, DDR crystals exposed to toluene,
N-methyl pyrrolidone, cyclohexyl amine, and butylamine did not show
any significant alteration in CO.sub.2 adsorption.
Alteration of the mass transfer kinetics was studied for samples
selectivated by DMDS, MDEA, and DMEA. The selectivation was
believed to be due in part to entry of barrier compounds into the
pores of the adsorbent, resulting in a reduced likelihood of
CH.sub.4 diffusion into the crystals. For illustration, a clear
example of the impact of a barrier compound on CH.sub.4 transport
was produced by selectivating a DDR zeolite adsorbent using MDEA.
In this Example, DDR zeolite crystals were selectivated by exposing
the DDR zeolite crystals for about one month at a temperature of
.about.100.degree. C. to a base gas mixture that was initially in
contact with a liquid pool of MDEA. The liquid was not in contact
with the adsorbent and served to saturate the base gas mixture with
MDEA. The amount of MDEA in the pool contained more than 50 times
the number of molecules required to saturate the base gas mixture.
The saturated MDEA vapor in this experiment had a the concentration
of MDEA in the gas phase more than 5000 times greater than
concentrations expected in field applications for swing adsorption
units that a processing natural gas.
FIG. 4 shows ZLC studies for DDR crystals after various treatments.
In FIG. 4, the "baseline" plot shows an example of the output from
a ZLC measurement without inserting an adsorbent material to
capture the methane. The fresh sample curve shows the measured
diffusion of methane for DDR crystals prior to exposure to a
selectivating agent. The MDEA exposed curve corresponds to DDR
crystals selectivated by MDEA as described above. As shown in FIG.
4, the diffusion of methane out of the selectivated DDR crystals
was initially lower than for the non-selectivated crystals.
However, the diffusion out of the selectivated DDR crystals then
stabilized to have a shallower slope than the non-selectivated
crystals.
FIG. 5 shows additional ZLC studies at different flow rates of
CH.sub.4 for the MDEA selectivated crystals described above. In
addition to showing the measured ZLC diffusion amounts, FIG. 5 also
shows a fit of the shallow slope portion of the curve to a "surface
barrier" model of diffusion. As shown in FIG. 5, the surface
barrier model provided a reasonable fit for the long tail of the
diffusion. This indicates that the MDEA exposure had an effect
similar to formation of a surface barrier on the DDR crystals.
In addition to modifying the diffusion characteristics for
CH.sub.4, the MDEA also appeared to modify the adsorption
characteristics for CO.sub.2. FIG. 6 shows adsorption isotherms for
CO.sub.2 for both non-selectivated and selectivated DDR crystals.
As shown in FIG. 6, selectivation with MDEA in the presence or
absence of water resulted in comparable levels of selectivation.
The DDR crystals exposed to MDEA showed an approximately 30%
reduction in the amount of adsorbed CO.sub.2, as compared to
non-selectivated crystals.
As a further study, the degree of selectivation of DDR crystals was
investigated using a cyclic selectivation experiment. In the
experiment, N.sub.2 was bubbled through a barrier compound or
foulant. The N.sub.2 flow was then passed into a cell containing
DDR crystals. This exposed the crystals to the barrier compound.
Periodically, the N.sub.2 flow was stopped and a gas containing
CO.sub.2, N.sub.2, and CH.sub.4 was introduced. The cycle period
for the apparatus ranged from about 1-20 seconds. The working
capacity or adsorption of the crystals over time was compared with
the initial adsorption of the crystals prior to exposure to the
barrier compound.
FIG. 7 shows the working capacity of the DDR crystals over time as
the crystals were cyclically exposed to n-hexane. As shown in FIG.
7, the working capacity of the crystals for N.sub.2 adsorption was
at least about 70% of the initial working capacity throughout the
procedure. For CO.sub.2, the working capacity was at least about
80% of the working capacity throughout the procedure. This shows
that the selectivation with hexane had a modest impact on CO.sub.2
and N.sub.2 adsorption during .about.1-20 second long cycles. This
was not surprising, as the equilibration time for CO.sub.2 and
N.sub.2 is on the order of 0.1 seconds, so the CO.sub.2 and N.sub.2
adsorption amounts appeared to approach equilibrium levels during
the cycles. By contrast, the amount of CH.sub.4 adsorbed after
starting the cyclic exposure to n-hexane was significantly lower.
As shown in FIG. 7, the amount of adsorbed CH.sub.4 for the
selectivated DDR crystals was less than about 20% of the adsorbed
amount for non-selectivated crystals. This demonstrates that the
n-hexane was apparently selectivating the DDR crystals, so that
CH.sub.4 adsorption was inhibited to a greater degree than CO.sub.2
or N.sub.2 adsorption.
Similar cyclic fouling or selectivation studies were also performed
using methanol, toluene, and water as the foulant or barrier
compound. Cyclic fouling was conducted by exposing crystals to
vapors of the foulant at an activity of 50% of saturation carried
in a .about.100.degree. C. nitrogen stream whose pressure is swung
from about 150 psig (about 1.0 MPag) to about 20 psig (about 130
kPag) every .about.10-50 seconds. In these additional cyclic
fouling studies, the working capacity for the DDR crystals did not
appear to be impacted in a substantial way by the exposure for
times greater than one week to methanol, toluene, or water.
The dramatic change in the CH.sub.4 working capacity with only a
modest change in the CO.sub.2 and N.sub.2 working capacities was
believed to be due to at least partial entry of the hexane into the
pore network of the adsorbent. The fact that hexane appeared to act
as a barrier compound was deduced from measurements of diffusion
coefficients at .about.30.degree. C. As shown in FIG. 8, the
diffusion coefficients for alkanes containing 3 or more carbons
were substantially smaller than the diffusion coefficients for
methane or ethane.
In the cyclic fouling experiments performed at .about.100.degree.
C. with n-hexane at .about.50% of its saturated vapor pressure, it
was believed that a strong adsorption front advanced a distance
into the DDR crystals less than (D*t).sup.1/2. Based on the low
value of the hexane diffusion coefficient, for a .about.350-hour
exposure, the maximum distance of penetration was about 150
Angstroms. Thus, it is believed that all of the transport changes
were due to a thin layer of adsorbed hexane within the pore network
of the DDR crystals near the surface. At the cycle time used to
probe the fouling, the kinetic selectivity of DDR appeared to be
enhanced by a factor of at least 5 for CO.sub.2/CH.sub.4
separations and by at least a factor of 4 for N.sub.2/CH.sub.4
separations. Because of a relatively long cycle time and the nature
of the experimental protocol, selectivity enhancements achievable
in a swing adsorption separation process are expected to be 5-10
times greater than this.
The impact of selectivation on kinetically controlled swing
adsorption processes run with DDR has been modeled. Modeling
results showed that, for bulk acid gas or N.sub.2 removal,
selectivation increased methane recovery by lowering the
concentration of methane in the CO.sub.2 or N.sub.2 rich reject
stream. Selectivation may also mitigate fouling of DDR by molecular
species such as amines, DMDS, ethylene glycol, triethylene glycol,
MeOH, carboxylic acids, and N-methylpyrrolidone. It is believed
that selectivation of the crystal surface can inhibit or minimize
the entry of foulants into the pores of the crystals.
Example 2
Selectivation of DDR Crystals by Shorter Term Exposure to High
Concentrations of Mono- and Di-Branched Alkanes Above 250.degree.
C.
The size of mono- and di-branched alkanes was such that they would
not have been expected to go into the DDR pores, based on the IZA
published dimension of a hard sphere that can pass through the
pores. Fundamental molecular simulations of flexibility of DDR
frameworks, however, indicated that the pore structure occasionally
flexes to a size where they might be able to enter. From the
simulation of framework flexibility, it was still not clear if they
would transport into the framework even at high temperatures.
To test whether mono- and di-branched alkanes could be diffused
into DDR to fill enough pores to form a selectivation layer, a high
temperature (.about.290.degree. C.) selectivation procedure was
chosen, because the diffusion of sterically hindered molecules into
DDR was expected to be a highly activated process that would occur
orders of magnitude faster at .about.290.degree. C. than at room
temperature (.about.20-25.degree. C.).
To test selectivation with mono- and di-branched alkanes, a
different batch of DDR was used than that discussed in Example 1.
This batch of DDR crystals had a Si/Al ratio >250 and was found
to have significantly faster diffusion coefficients for CH.sub.4
than those discussed in Example 1. Crystals in the batch were
shaped like disks that were approximately uniform in size. The
2-dimensional connected pore structure allowing transport of
CO.sub.2 and CH.sub.4 molecules was believed to be in the plane of
the disk. A characteristic dimension for transport in the plane of
the disk was about 7-9 microns.
To selectivate, quarter gram samples of the as-synthesized DDR
crystals were loaded into a .about.1/8-inch diameter pressure cell
connected to a pressurized reservoir that was filled with 25 grams
of a liquid phase mono- or di-branched alkane chosen for the
selectivation. Purity of the liquids used in these experiments was
greater than 99%. To selectivate, the cell holding the zeolite was
filled with the liquid, pressurized to about 700 psig (about 4.8
MPag) and then heated to .about.290.degree. C. Once the samples had
been held at temperature for the targeted time period, the pressure
cell was rapidly cooled, and the selectivated DDR powder was
recovered. Before conducting transport or TGA studies, excess
molecules were removed by placing the powder in a nitrogen purged
oven at .about.70.degree. C. for at least .about.2 hours and then
by placing the sample in a vacuum oven set in a range between
.about.60.degree. C. and .about.70.degree. C. for .about.12
hours.
To determine if a sufficient quantity of the selectivating
molecules were trapped in the DDR, the weight percent of the
selectivating molecule incorporated into the zeolite was assessed
with TGA measurements. In the TGA, milligram quantities of
selectivated powders were heated in a flowing nitrogen atmosphere
to .about.120.degree. C. to remove species such as water and
CO.sub.2 that can adsorb in the sample. Samples were then held at
.about.120.degree. C. for approximately an hour to make sure that
the sample weight was stable and that nothing else continued to
desorb. Temperature was then ramped at .about.6.degree. C./min to a
temperature of .about.600.degree. C., held at that temperature for
.about.1 hour, and then rapidly cooled back to .about.120.degree.
C. Differences in the weight recorded at .about.120.degree. C.
before and after the temperature ramp were used to assess the
weight loss that occurred when the sample was heated and held at
.about.600.degree. C. This protocol was believed to eliminate
errors from buoyancy effects in the TGA. The table below shows the
weight lost in the TGA for different molecular selectivation of the
DDR powder. This weight loss appeared to come from both desorption
of the selectivating molecules at high temperature and from weight
loss associated with changes in defects such as hydroxyls in the
DDR crystalline lattice. To more accurately assess the uptake of
selectivating molecules, a sample was treated with no selectivating
molecules present. The weight loss of this sample was used to
identify the weight loss due with changes in defects such as
hydroxyls in the DDR crystalline lattice. Using this as a
reference, Table 1 below shows the weight uptake (in wt %) of the
selectivating molecules into the DDR. If the DDR lattice was
completely filled with any of these molecules, then the weight
change would have been greater than 5 wt %.
TABLE-US-00001 TABLE 1 Weight Loading of Exposure Lost in
Selectivating Selectivating Temp Pressure Time TGA Molecule
Molecule (.degree. C.) (psig) (hrs) (wt %) (wt %) Treatment with no
290 700 40 0.3 0.0 selectivating molecule 2-Methylpentane 290 700
40 0.5 0.2 3-methylpentane 290 700 22 0.7 0.4 2,5-Dimethylhexane
290 700 20 0.6 0.4 3-methylhexane 290 700 20 0.7 0.4 2-methylhexane
290 700 20 0.5 0.2 2,4- 290 700 20 0.5 0.2 Dimethylpentane 2,3- 290
700 20 0.6 0.3 Dimethylpentane
To demonstrate that these modest loading levels of the
selectivating molecules can significantly reduce the methane
diffusivity, ZLC experiments were conducted on three of the
selectivated samples and on the as-synthesized DDR. The shape of
the ZLC curve for the as-synthesized material did not conform to
the expected theoretical shape and fits to different sections of
the curve yielded CH.sub.4 diffusivities at .about.30.degree. C.
ranging from about 8.times.10.sup.-13 m.sup.2/s to about
50.times.10.sup.-13 m.sup.2/s. Fits to these sections of the curve
were interpreted as a description of how molecules at different
loadings appear to transport out of the zeolite. A loading-averaged
CH.sub.4 diffusivity of these fits was taken to be
.about.15.times.10.sup.-13 m.sup.2/s. Single component PSA studies
of a film made from this zeolite supported a higher average
diffusivity in the range of .about.30.times.10.sup.-13 m.sup.2/s at
the PSA testing condition of .about.10 bar (.about.10.1 MPag) and
.about.22.degree. C. For all three selectivated samples, the shape
of the ZLC curve matched the expected theoretical form. The sample
selectivated with 2-methyl pentane had a CH.sub.4 diffusivity at
.about.30.degree. C. of .about.2.2.times.10.sup.-13 m.sup.2/s.
Repeat measurements of the DDR sample selectivated with 3-methyl
pentane yielded CH.sub.4 diffusivities at .about.30.degree. C. of
.about.2.2.times.10.sup.-13 m.sup.2/s and
.about.2.5.times.10.sup.-13 m.sup.2/s. Repeat measurements of the
DDR sample selectivated with 2,5 di-methyl hexane yielded CH.sub.4
diffusivities at .about.30.degree. C. of about 1.6.times.10.sup.-13
m.sup.2/s and about 1.9.times.10.sup.-13 m.sup.2/s.
Example 3
Selectivation of a DDR Films by Shorter Term Exposure to High
Concentrations of a Mono Branched Alkanes at Temperatures Above
250.degree. C.
Example 3 involves exposing an adsorbent to the barrier compound
after the adsorbent has been incorporated into a bed or a component
that can be assembled into a bed and used in the contactor within a
swing adsorption unit.
This Example used a batch of DDR crystals different from the
batches in Examples 1 and 2, in that this batch of DDR crystals had
a broader particle size distribution. This batch of DDR crystals
also had a characteristic dimension of .about.13 .mu.m and a
CH.sub.4 diffusivity measured by ZLC at .about.30.degree. C. of
.about.12.times.10.sup.-13 m.sup.2/s.
Approximately 200 micron thick films were cast from a slurry of a
colloidal silica binder and DDR crystals using a doctor blade to
apply them to a flat polished steel plate that had been previously
coated with a thin (<25 micron) zirconia layer to improve
adhesion. Once applied, the plate was heated to .about.150.degree.
C. to cure the coating. It was estimated that open pores occupied
less than about 30 vol % of the cured film, and the mass of DDR was
>15 times that of the binder. The steel plate was .about.1.9
inches wide and .about.5 inches long. In the blade process, the
edges of the plate were masked so that the width of the film on the
plate was .about.1.6 inches. The uncoated edges of the plate were
used to seal the plate into a cell that could be installed in a
single complement pressure swing adsorption unit. Once the plate
was sealed into the cell, the flow channel that was left on top of
the film was approximately 250 microns thick.
Single component pressure swing adsorption experiments were first
performed at room temperature (.about.20-25.degree. C.) and
.about.50.degree. C. to characterize the transport properties of
the as synthesized film. Individual experiments were conducted at
each temperature with CO.sub.2, CH.sub.4, N.sub.2, and He. In each
experiment, the adsorbent was cyclically pressurized to about 125
psig (about 863 kPag) and depressurized to about 25 psig (about 170
kPag). Measurements were taken after a cyclic steady state
operation was achieved. Pressurization was done by opening a valve
to connect a tank pressurized to .about.150 psig (.about.1.0 MPag)
and filled with a packing (mitigating temperature changes) to the
cell holding the adsorbent at a pressure of .about.25 psig
(.about.170 kPag). After about 0.25 seconds, the valve was closed,
at which time the tank pressure fell to about 125 psig (about 863
kPag), and the cell pressure rose to about 125 psig (about 863
kPag). Accurate measurements of the real rise and fall of the tank
and cell pressures and temperatures were used in conjunction with
measurements of the behavior of the system with a non-adsorbing gas
(i.e., He) to deduce the number of moles transferred and the number
of moles taken up by the adsorbent. The cell was sealed off for
.about.30 seconds to monitor any slow adsorption processes that
could produce time dependent changes in the number of moles taken
up by the adsorbent. Accurate measurements of changes in sealed
cell pressure and temperature were used in conjunction with
measurements of the behavior of the system with a non-adsorbing gas
(i.e., He) to quantify time dependent loading changes. The LDF
approximation was then used to convert any observed change into an
effective diffusion coefficient. The shortest time scale on which
this type of analysis can be conducted is .about.0.25-0.5 seconds.
This provided an upper bound on the effective diffusion coefficient
that can be measured. After holding the cell in a sealed state for
.about.30 seconds, the valve to the depressurization tank initially
at .about.1 psig (.about.7 kPag) was opened for .about.10 seconds.
During this time, the pressure in cell and depressurization tank
equilibrated to about 25 psig (about 170 kPag). After the
.about.10-second long depressurization, the valve between the
depressurization tank and cell was closed, and the cell was sealed
off for approximately 10 more seconds while waiting to start a new
cycle. Each subsequent cycle had the same valve timing as the
first. Before the start of each cycle, the pressures in the feed
and depressurization thanks were reset.
In the PSA measurements, CO.sub.2 and N.sub.2 appeared to be fully
equilibrated in less than 0.25 seconds. As such, their diffusion
coefficient was too fast to measure, providing a lower bound of
.about.2 10.sup.-11 m.sup.2/s for the effective CO.sub.2 and
N.sub.2 diffusion coefficients in the DDR crystals in the
as-synthesized film. With CH.sub.4, the pressure changed in the
cell in a manner that could be approximately fit using a LDF mass
transfer coefficient. The effective CH.sub.4 diffusion coefficient
extracted from the LDF of the room temperature experiment was
approximately 35.times.10.sup.-13 m.sup.2/s. This was approximately
three times larger than the ZLC measurement at .about.30.degree. C.
ZLC is a low pressure measurement, and it is known that effective
diffusion coefficients can be pressure dependent.
After measuring diffusion coefficients in the as-synthesized film,
the film was removed from the single complement PSA unit and
selectivated with 3-methyl pentane using a procedure very similar
to that described in Example 2. The selectivation was done at
.about.700 psig (.about.4.8 MPag) and .about.290.degree. C. for
.about.96 hours in an autoclave that was large enough to contain
the steel plate.
After selectivating, the film was reloaded into the single
complement PSA unit and remeasured. It was found that CO.sub.2 and
N.sub.2 fully equilibrated in less than 0.25 seconds in a manner
that was identical to the as-synthesized film. As such, there was
no detectable change in the kinetics for CO.sub.2 and N.sub.2
diffusion coefficients from the selectivation process. The methane
diffusion coefficient determined from the LDF fits dropped by
factor of about 4.2 at both room temperature and 50.degree. C. At
room temperature, the effective CH.sub.4 diffusion coefficient in
the 3-methyl pentane selectivated DDR was approximately
8.times.10.sup.-13 m.sup.2/s.
Example 4
Selectivation of DDR Crystals by Shorter Term Exposure to High
Concentrations of a Linear Alkanes at Temperatures Above
250.degree. C.
This Example used the DDR material and selectivation process of
Example 2. It was found that selectivating with n-octane at
.about.700 psig (.about.4.8 MPag) and .about.290.degree. C. for
.about.3 hours produced a loading of the n-octane greater than 4 wt
% in the DDR crystals. This loading was greater than .about.75% of
the expected q.sub.s for n-octane and was undesirably large.
Similar results were found for selectivation with n-hexane and
n-heptane for exposure times longer than .about.2 hours. To
demonstrate that the loading could be reduced by shortening the
exposure time, selectivation was performed with n-hexane at
.about.700 psig (.about.4.8 MPag) and .about.290.degree. C. for
about .about.0.75 hours. After the .about.0.75 hours selectivation,
it was found that the loading was .about.1.25 wt %. ZLC
measurements were performed on the sample, and it was found that
the methane diffusivity had fallen below a level that could be
detected by ZLC. The ZLC curve closely matched the instrumental
resolution, and it was estimated that the CH.sub.4 diffusivity had
fallen below .about.10.sup.-14 m.sup.2/s.
Example 5
Selectivation of DDR Crystals by Shorter Term Exposure to High
Concentrations of a Methyl Branched Primary Alcohols at Above
250.degree. C.
This Example used the DDR material and selectivation process of
Example 2. Similar to Example 2, it was not a priori expected that
a methyl branched primary alcohol could diffuse into DDR.
TGA studies of the uptake of methyl branched primary alcohols in
selectivated DDR crystals are summarized in Table 2 below. In all
of these TGA studies, the predominant weight loss occurred near
.about.400.degree. C. This indicated some type of chemical bonding
between the DDR and the methyl branched primary alcohols. It was
hypothesized that the methyl branched primary alcohols had reacted
with defects associated with hydroxyls and other defects in the DDR
crystals. NMR studies of DDR and methyl branched primary alcohol
selectivated DDR supported this hypothesis. Such a reaction
enhances the stability of the selectivation.
TABLE-US-00002 TABLE 2 Weight Loading of Exposure Lost
Selectivating Selectivating Temp Pressure Time in TGA Molecule
Molecule (.degree. C.) (psig) (hrs) (wt %) (wt %) Treatment with no
290 700 40 0.3 0.0 selectivating molecule 2-methyl-1-propanol 270
700 14 0.9 0.6 2-methyl-1-pentanol 270 700 15 1.5 1.2
2-methyl-1-butanol 270 700 14 1.3 1.0
To determine the impact on CO.sub.2 and CH.sub.4 transport from the
methyl branched primary alcohol selectivation, a ballistic
chromatographic procedure was used. In this method .about.5-15 mg
of adsorbent was loaded into a cell .about.1 mm in diameter and
.about.1.5 cm long. The cell was then loaded into a chromatographic
unit in a manner such that the connections to the cell did not
cause any significant hydrodynamic back mixing when gases were
flowed through the cell at flow rates of .about.5-40 SCCM and
pressures ranging from .about.1 barg (.about.100 kPag) to .about.5
barg (.about.500 kPag). To be able to interpret data, flow though
the cell had to be essentially assumed to be "plug flow". Before
beginning measurements, the adsorbent was conditioned between
.about.70.degree. C. and .about.225.degree. C. in either flowing
helium or flowing nitrogen.
To determine transport, a mass spectrometer having a time response
faster than .about.0.25 seconds was used to measure changes of
either CO.sub.2 or CH.sub.4 concentrations flowing out of the cell.
Gases used to probe the transport were mixtures of .about.1-20%
CH.sub.4 in He or .about.1-20% CO.sub.2 in He. In this Example, the
probe gasses were .about.10% CH.sub.4 in He and .about.10% CO.sub.2
in He. At the start of an experiment, He was flowed through the
cell at either .about.5 cm.sup.3/min or .about.25 cm.sup.3/min. A
valve was thrown, and the probe gas was flowed through the cell at
the same rate as the He. After a prescribed period of time, the
valve was thrown again, and helium was flowed through the cell at
the original flow rate. In the design of the equipment, it was
important that the velocity in the inlet streams remained
approximately constant when the valve was thrown. There were
several methods to do this, one of which being to have the inlet
valve switch flow from a cell to a "dummy" path having about the
same hydrodynamic resistance as the cell. In such a case, the He
and probe gas would always be flowing and would only be switched
from flowing through the cell and flowing through to the "dummy"
flow path.
Much of the analysis of transport for CH.sub.4 came from the
shortest .about.6-second time during which the probe gas flowed
through the cell. Other lengths of time studied for the probe gas
flow through the sample were .about.30 seconds, .about.1 minute,
and .about.2 minutes. Difference between results at .about.5
cm.sup.3/min and .about.25 cm.sup.3/min were used to assess whether
CO.sub.2 was fully equilibrated with the adsorbent before it passed
through the cell. Analysis of data also relied on recording the
time response of a blank cell.
Transport through the 2-methyl-1-propanol selectivated sample was
compared with that of the base DDR using ballistic chromatography.
For CO.sub.2, there was a distinct breakthrough that was very
similar in the 2-methyl-1-propanol selectivated sample and the base
DDR. The shape and position of the front scaled with flow rate in a
manner that indicated that CO.sub.2 had fully equilibrated with
samples even at the highest flow rates. As expected for full
equilibration, changing the time for which gas was flowed through
the cell from .about.6 seconds to .about.30 seconds did not appear
to change the nature of the breakthrough front. This indicated that
the CO.sub.2 diffusion coefficients in the selectivated and base
DDR samples were greater than 10.sup.-10 m.sup.2/s. Comparing the
CH.sub.4 measurements of the selectivated and base DDR samples
provided an estimate of the change in CH.sub.4 diffusion from
selectivation. Both modeling and direct comparison indicated that
the 2-methyl-1-propanol selectivation reduced the effective
CH.sub.4 diffusion coefficient by at least a factor of 2.5.
Similar conclusions were made about the impact of
2-methyl-1-propanol selectivation of DDR films using the methods
and materials of Example 3. Films that were selectivated with
2-methyl-1-propanol at .about.700 psig (.about.4.8 MPag) and
.about.290.degree. C. for .about.44 hours had a .about.3-fold
reduction in the CH.sub.4 diffusion coefficient with no significant
impact on the observed kinetics for CO.sub.2.
Example 6
Selectivation of DDR Crystals by Shorter Term Exposure to High
Concentrations of Secondary Alcohols at Temperatures Above
250.degree. C.
This Example used the DDR material and selectivation process of
Example 2. TGA studies of the uptake of secondary alcohols in
selectivated DDR crystals are summarized below in Table 3. Similar
to the selectivation with methyl branched primary alcohols, the
predominant weight loss occurred near .about.400.degree. C. This
indicated some type of chemical bonding between the DDR and the
secondary alcohols. It was hypothesized that the secondary alcohols
reacted with defects associated with hydroxyls and other defects in
the DDR crystals. Such a reaction appeared to enhance the stability
of the selectivation.
TABLE-US-00003 TABLE 3 Weight Loading of Exposure Lost in
Selectivating Selectivating Temp Pressure Time TGA Molecule
Molecule (.degree. C.) (psig) (hrs) (wt %) (wt %) Treatment with no
290 700 40 0.3 0.0 selectivating molecule 2-propanol 270 700 15 0.7
0.4 2-butanol 270 700 14 1.8 1.5 2-pentanol 270 700 14 0.5 0.2
2-hexanol 270 700 14 0.6 0.3 2-heptanol 270 700 14 0.7 0.4
Transport through the 2-hexanol selectivated sample was compared
with that of the base DDR using ballistic chromatography. Studies
indicated that the effective CH.sub.4 diffusion coefficient fell by
more than a factor of 10, while the CO.sub.2 transport appeared to
exhibit almost no change.
Example 7
Selectivation of DDR Films by Shorter Term Exposure to High
Concentrations of Secondary Alcohols at Temperatures Above
250.degree. C.
This Example used the DDR material and selectivation process of
Example 3. Single component PSA measurements were used to assess
transport in a .about.250 micron thick DDR film before and after
selectivating for .about.44 hours at .about.700 psig (.about.4.8
MPag) and .about.290.degree. C. with 2-hexanol. Comparison of
CO.sub.2 measurements taken before and after selectivation showed
no detectable change in the CO.sub.2 kinetics. Analysis of the PSA
response placed a lower bound of .about.2.times.10.sup.-11
m.sup.2/s on the effective CO.sub.2 diffusion coefficient in the
unselectivated and the 2-hexanol selectivated DDR film. With
CH.sub.4, the pressure changed in the cell in a manner that could
be approximately fit using a LDF mass transfer coefficient. The
effective CH.sub.4 diffusion coefficient for DDR extracted from the
LDF analysis of the room temperature experiment on the 2-hexanol
selectivated DDR film was approximately 2.times.10.sup.-13
m.sup.2/s. LDF analysis showed that the effective CH.sub.4
diffusion coefficient in the 2-hexanol selectivated DDR film was
.about.15 times smaller than that in the unselectivated film at
both room temperature and 50.degree. C.
Additional single component PSA measurements were used to assess
transport in a .about.250 micron thick DDR film before and after
selectivating for .about.14 hours at .about.700 psig (.about.4.8
MPag) and .about.290.degree. C. with 2-octanol. Comparison of
CO.sub.2 measurements taken at room temperature before and after
selectivation showed no detectable change in the CO.sub.2 kinetics.
Analysis of the PSA response placed a lower bound of
.about.2.times.10.sup.-11 m.sup.2/s on the effective CO.sub.2
diffusion coefficient in the unselectivated and the 2-octanol
selectivated DDR film. With CH.sub.4, the pressure changed in the
cell in a manner that could be approximately fit using a LDF mass
transfer coefficient. The effective CH.sub.4 diffusion coefficient
for DDR extracted from the LDF analysis of the room temperature
experiment on the 2-octanol selectivated DDR film was
.about.0.3.times.10.sup.-13 m.sup.2/s. LDF analysis showed that the
effective CH.sub.4 diffusion coefficient in the 2-octanol
selectivated DDR film was more than 100 times smaller than that in
the unselectivated film.
Although the present invention has been described in terms of
specific embodiments, it is not so limited. Suitable
alterations/modifications for operation under specific conditions
should be apparent to those skilled in the art. It is therefore
intended that the following claims be interpreted as covering all
such alterations/modifications as fall within the true spirit/scope
of the invention.
* * * * *
References